The behavioural causes of bullwhip effect in supply chains: a

systematic literature review

Y. Yang,1 J. Lin,1 G. Liu,1 L. Zhou2*

1
International Business School Suzhou, Xi’an Jiaotong-Liverpool University, PR China

[email protected]; [email protected]; [email protected]

2
Faculty of Business, University of Greenwich, UK

[email protected]

Abstract

The bullwhip effect, also known as demand information amplification, is one of the principal

obstacles in supply chains. In recent decades, extensive studies have explored its operational

causes and have proposed corresponding solutions in the context of production inventory and

supply chain systems. However, the underlying assumption of these studies is that human

decision-making is always rational. Yet, this is not always the case, and an increasing number

of recent studies have argued that behavioural and psychological factors play a key role in

generating the bullwhip effect in real-world supply chains. Given the prevalence of such

research, the main objective of this study is to provide a systematic literature review on the

bullwhip effect from the behavioural operations perspective. Using databases, including

Scopus, Wiley Online Library, Google Scholar and Science Direct, we selected, summarised

and analysed 53 academic studies. We find that most studies build their models and

simulations based on the ‘beer distribution game’ and analyse the results at the individual

1
level. We also demonstrate the importance of studying human factors in the bullwhip effect

through adapting Sterman’s double-loop learning model. Based on this model, we categorise

and analyse the behavioural factors that have been studied and identify the explored

behavioural factors for future research. Based on our findings, we suggest that future studies

could consider social and cultural influences on decision-making in studying the bullwhip

effect. In addition, further aspects of human mental models that cause this effect can be

explored.

Keywords: bullwhip effect, behavioural causes, systematic review, cognitive psychology,

supply chains

2
1. Introduction

The bullwhip effect, also known as demand information amplification or the Forrester

(1958, 1961) effect, is a phenomenon whereby a small variation in end-customer demand

leads to a significant fluctuation in orders that the upstream supplier receives in the supply

chain system (Lee, Padmanabhan, & Whang 1997). This phenomenon has a tremendous

negative influence on supply chain performance, with associated costs, such as machine

capacity and staff recruitment fluctuations and excessive inventory levels.

Traditional studies on the bullwhip effect have focused on the operational perspective,

including its causes (Lee, Padmanabhan, & Whang 1997; Lin et al. 2017) and mitigation

solutions, such as reducing lead time and increasing information transparency (Lee,

Padmanabhan, & Whang 1997; Wang & Disney 2016). However, human factors—

psychological or behavioural causes—may lead to the bullwhip effect (Sterman 2006),

because irrational decisions can be produced by individual cognitive limitations (Gino &

Pisano 2008; Loch & Wu 2005) and stressful environments (Sterman & Dogan 2015). Thus,

even under a rational operational decision-making process (e.g. when the order decision is

mathematically optimal), irrational behaviours still strongly influence the bullwhip effect

level in supply chain systems (Sterman 1989a).

An increasing number of studies have explored the behavioural causes of the bullwhip

effect in supply chains over the past three decades. The topics studied include inventory

information sharing (e.g. Croson & Donohue 2003, 2006), training and communication (Wu

& Katok 2006), reactions to reverse bullwhip effect and supply shock (Rong, Shen, & Snyder

2008), trust in collaboration (Cao, Baker, & Schniederjans 2014) and human judgement in

forecasting (Baecke, De Baets, & Vanderheyden 2017). These studies integrated human

3
behaviours into research on operations management and formed a new research approach to

the bullwhip effect in supply chains by adopting theories from cognitive and social

psychology.

Given the prevalence of behavioural research on the bullwhip effect, we aimed to

systematically review relevant works. The systematic review method is considered an

evidence-based practice (Jones & Gatrell 2014), which has been expanding driven by topic-

related questions in different disciplines, and its application and expansion have enriched and

developed knowledge bases and research methods in different fields (Tranfield, Denyer, &

Smart 2003). We acknowledge that several reviews have been devoted to the bullwhip effect

(Bhattacharya & Bandyopadhyay 2011; Geary, Disney, & Towill 2006; Giard & Sali 2013;

Miragliotta 2006; Wang & Disney 2016), which have comprehensively reviewed the methods

applied for modelling and reducing bullwhip effect. For example, Wang and Disney (2016)

classified methods, particular the operational research approaches applied in modelling

bullwhip effect, and highlighted solutions, such as the value of information sharing across the

supply chains (Giard & Sali 2013). Further, recent reviews (Arvan et al. 2019; Fahimnia,

Sander, & Siemsen 2020; Perera et al. 2019) contributed to revealing the scope of human

judgement in supply chain forecasting, which could be considered as one of behavioural

factors driving the bullwhip effect. Within the board operations and supply chain context,

Fahimnia et al. (2019) reviewed and categorised 12 operations contexts as well as emerging

topic considerations, and Perera, Fahimnia and Tokar (2020) reviewed the behavioural

experiments in ordering and inventory decisions.

Nevertheless, most such reviews have only considered behavioural factors for the

bullwhip effect as one of the reviewed topics, or have reviewed the behavioural factors for

bullwhip effect associated with operational causes (e.g. forecasting). Further, some reviews

4
(e.g. Bhattacharya & Bandyopadhyay 2011; Geary, Disney, & Towill 2006; Miragliotta 2006)

are outdated. To the best of our knowledge, until date, no review has systematically explored

the behavioural causes of the bullwhip effect within the context of supply chains. Therefore,

the main objective of this study was to classify, analyse and synthesise the behavioural causes

of the bullwhip effect from different aspects. To this end, we summarised the research

methods, research levels and research clusters in the study of behavioural factors of the

bullwhip effect, and we categorised these behavioural factors. This study offers two main

contributions, as follows:

1. demonstrating that human and behavioural factors cannot be ignored in examining the

bullwhip effect, which will motivate future studies to include human factors in the

decision-making process of the bullwhip effect

2. identifying research gaps to indicate a future research agenda through understanding

the influence of human and behavioural factors on the bullwhip effect and considering

possible problems in real business scenarios.

The remainder of this paper is organised as follows. Section 2 presents the relevant

background and fundamental theories of this topic. Section 3 includes the method adopted in

this study, and Section 4 mainly discusses the aggregated results found in the collected

studies. Section 5 presents the content classification and analysis, and Section 6 emphasises

the existing research gaps based on the analysis results, proposes potential future research

directions and discusses the limitations of this study.

2. Background and Fundamental Theories

2.1 Brief History of Bullwhip Effect Research

5
 The concept of the ‘bullwhip effect’ was discovered and termed ‘demand amplification’ by

Professor Jay Forrester (1958, 1961) during the development of the system dynamics

discipline. Burbidge (1961, 1984) proposed a method from the operational perspective to

control inventories and asserted that the bullwhip effect was caused by using stock control

ordering during the process of transmitting demand. Sterman (1989a, 1989b) used the ‘beer

distribution game’ experiment to explore the bullwhip effect and indicated that it was caused

by irrational decisions resulting from decision-makers’ misperceptions of feedback. Further,

Lee, Padmanabhan and Whang (1997) proposed four underlying causes for bullwhip effect

based on a comprehensive systematic analysis: order batching, the shortage and rationing

gaming, demand signal processing and fluctuating prices. Based on these causes, more

scholars began to study the bullwhip effect—for instance, Cachon and Lariviere (1999)

examined rationing gaming and order batching by using a supply chain system with N-

retailers and one supplier, and Kim et al. (2006) introduced stochastic lead time.

Such studies focusing on the operational causes of the bullwhip effect have assumed that

humans are always rational. However, a series of experimental studies have indicated that

human rationality is bounded, and hence, individuals’ decisions can deviate from the expected

decision and subsequently affect performance. Thus, the behavioural operations approach has

emerged and provides a new theoretical basis for exploring the bullwhip effect further.

6
2.2 Behavioural Operations Research

The study of the bullwhip effect from the perspective of behaviour can be traced to

Forrester (1958). On this basis, Sterman (1989a) used experimental methods to explore the

behavioural causes of the bullwhip effect and asserted that the behaviour of the people

managing the supply chain system is key to improving its performance. Thus, it is necessary

to further explore the behavioural factors of the bullwhip effect. Moreover, in recent years,

scholars have proposed that operations management be examined from the behavioural

operations perspective, which provides a new research approach to the bullwhip effect. For

example, Loch and Wu (2005) discussed how to consider more practical behavioural factors

in theoretical analysis models and proposed that human behaviours result from individual

cognitive limitations, social interactions with others and cultural evolution and transmission.

Further, Bendoly, Donohue and Schultz (2006) and Bendoly et al. (2010) summarised

relevant psychological theories and reviewed experimental research and knowledge to reveal

the influence of human behaviour on operations management. They assumed that knowledge

on behavioural operations mainly derives from system dynamics, group dynamics, social

psychology and cognitive psychology. In addition, Gino and Pisano (2008) stated that in

investigating behavioural operation, a new approach is to study operations management by

combining cognitive and social psychology theories. This approach involves exploring the

relevant attributes of cognition, the dynamics of groups and organisations that affect

operations and the interactions between these attributes and operation systems.

Given that most operations occur in the context of large organisations, communication,

learning and culture are crucial. However, it is also necessary to consider elements of

cognitive psychology, which encompasses mental psychological processes, involving topics

on decision-making, emotion, perception, memory and problem-solving (Gino & Pisano

7
2008). To conclude, explorations of the causes of the bullwhip effect from the behavioural

perspective are significant to the supply chain management field, given that the behaviours of

decision-makers who manage the supply chain have marked effects on supply chain

performance. Therefore, the analysis of human cognitive and social factors from a

psychological perspective is essential in investigating the bullwhip effect.

3. Methodology

This study adopted a systematic review method that has four stages, namely, the

identification, selection and evaluation of sources followed by the analysis of data (e.g.

Bryman 2012; Hart 1998). A systematic literature review is aimed at minimising selection

bias during the process of collecting relevant articles by defining and excluding keywords to

describe and discuss the search results (Saunders, Lewis, & Thornhill 2009; Spina et al.

2013). Compared with the traditional review method, its process and results are more

scientific, objective and transparent (Durach, Kembro, & Wieland 2017; Tranfield, Denyer, &

Smart 2003). The details of this method are discussed in the following subsections, and the

literature data collection and screening process is displayed in Figure 1.

8
 Figure 1: Literature data collection and screening process

3.1 Identification of Keywords and Sources

The keywords were determined to indicate the two main spheres that comprise this

literature review—the bullwhip effect and the associated psychological factors. All the co-

authors discussed and agreed on all the search terms, and the fourth author was the gatekeeper

of this study. The study was based on a keyword search in the Scopus database. We accessed

this database using the search field ‘article title, abstract and keywords’ and identified the

9
specific search terms to ensure that we collected all relevant documents. We combined the

keywords ‘bullwhip effect’, ‘beer game’ and ‘beer distribution game’ with each of the

following search terms: ‘reaction’, ‘preference’, ‘misperception’, ‘perception’, ‘individual’,

‘human’, ‘behaviour’, ‘cognitive’, ‘heuristics’, ‘social’, ‘interaction’, ‘learning’,

‘communication’, ‘judgement’, ‘psychology’, ‘behavioural operations’, ‘decision’,

‘rationality’, ‘trust’, ‘problem solving’, ‘personality’ and ‘overconfidence’. The keywords

‘beer game’ and ‘beer distribution game’ were also used in the search because this

experimental method is widely used to study the bullwhip effect and the decision-making

process in a dynamic environment. Moreover, the search results for ‘bullwhip effect’ were

limited; thus, ‘beer game’ and ‘beer distribution game’ were added to expand the search range

of this research topic.

We found a total of 586 articles from the databases of Scopus, Science Direct and Wiley

Online Library. To enhance rigour, we used Google Scholar to cross-check after completing

source identification and added 32 new articles after cross-checking, resulting in a list of 618

articles. To align with the scope, we limited the results to peer-reviewed articles published in

all date ranges and in English, since peer-reviewed articles are the most widely used

knowledge base for validating results and publishing new findings in the research community

(Spina et al. 2013). We removed articles that were not published in English and grey

literature, such as books, book chapters, conference papers, reviews, notes and dissertations

(Bryman 2012; Eksoz, Mansouri, & Bourlakis 2014; Perera et al. 2019). Thus, the final list

had 383 journal articles published until January 2021.

3.2 Source Selection

10
 Source selection includes investigating the data and corresponding sources based on the

review aims (Spina et al. 2013). In systematic literature reviews, certain keywords are

excluded to refine the search results to match the boundaries systematically (Saunders, Lewis,

& Thornhill 2009). Thus, in the present study, we excluded keywords comprising terms that

we considered beyond the scope of this topic and unrelated to the areas of human factors

associated with the bullwhip effect. For instance, we excluded ‘marketing’, ‘Markov

processes’, ‘algorithms’, ‘genetic algorithms’, ‘mean square error’, ‘Lyapunov methods’ and

‘production engineering’. This method reduced the number of articles in the peer-reviewed

literature pool; however, we carefully checked every article we removed to ensure that it was

indeed beyond the research scope.

We checked the remaining 139 articles one by one to ensure that they were consistent with

our research scope and excluded a further 86 from the list, which we considered beyond the

research scope. These excluded articles treated the supply chain as a ‘hard’ system and used

operations research–based methods to explore the dynamic behaviour of the bullwhip effect,

such as studies on control engineering (e.g. Hofmann 2017; Naim et al. 2017), statistical

approaches (e.g. Costantino et al. 2014; Disney, Towill, & Van De Velde 2004), hard

simulation (e.g. Liu, Howley & Duggan 2012; Poornikoo & Qureshi 2019) and mathematical

optimisation (e.g. Fu et al. 2014; Sadeghi et al. 2013). As a result, 53 peer-reviewed studies,

published from 1988 to January 2021 were selected as the final literature pool. This literature

pool was shared among, and confirmed by, all co-authors.

3.3 Source Evaluation

The purpose of source evaluation is to extract data by coding and categorising the studies

selected. We designed the coding process to understand the approach, focus and findings of

11
the selected studies on the bullwhip effect. Analysing the information these studies contained

was essential to determining the directions of the literature from the aspects of the research

areas considered over time in a systematic review (Snyder 2019). Table 1 displays the coding

criteria we used, including the coding items, their description and the reason for using these

items. The results of the coding were recorded in Microsoft Excel tables for data extraction

and data analysis.

Code Description Reason for Use

Authors’ names Authors of the study To identify studies by authors
Publication year Year in which study was published To enable a view of studies over time
Journal Journal of final publication To identify trends in published
journal articles
Title Title of the study To ensure the title of each article was
correct
Research The methods used in reviewed studies: To identify trends in methods
methods - modelling
- simulation experiment (simulating a business
scenario that involved subjects without
control and experimental groups)
- controlled experiment
- case studies
- survey/questionnaire
- computer simulation (mathematical
simulation)
Research area Topic and research area on which the study To identify the main areas of interest
cluster focused addressed in the studies
Identification Number assigned to each article To ensure all articles were coded
number
Research Analysis level of the study: To identify trends in analysis level
analysis level - individual level
- group and organisational level
- cultural level
Findings Influence on bullwhip effect and on supply chain To investigate the findings of each
article and the effects of the findings
on the research objectives
Behavioural Types of behavioural factors studied in bullwhip To identify the behavioural factors
factors effect research: that the articles studied
- social interaction (communication and
coordination, information sharing and trust)
- information feedback (perception, ambiguity
and debiasing)
- mental models (judgement anchoring and
adjustment, framing effect, overconfidence,

12
 underweighting and overweighting,
rationality/bounded rationality, misperception
of feedback, procedural rationality, problem-
solving, system thinking and cognitive
reflection)
- strategy (decision patterns and risk aversion)
- behaviour (hoarding behaviour and inaccurate
behaviour)
- individual traits (personality, and age and
experience)
- emotions (frustration and helplessness, and
panic)

Table 1: Coding criteria for the systematic review

3.4 Data Analysis

The process of data analysis in a systematic review involves extracting and synthesising

the relevant information from the selected studies and analysing the results to enable the

reviewer to identify future research directions (Snyder 2019). Therefore, we analysed all the

data extracted from the final list and used the analysis results as the basis for determining

research gaps. The aggregate results are presented in Section 4. Moreover, given that we

aimed to analyse the current situation of the studies on the bullwhip effect involving human

behavioural factors, we considered it necessary to classify the behavioural factors of the

selected articles and discuss the results.

We adapted Sterman’s (1994, 2000) double-loop learning model to categorise the content

of the selected articles (Figure 2). This model includes three feedback loops in the learning

and decision-making process. The first feedback loop involves ‘real world’, ‘information

feedback’ and ‘decisions’. In this loop, humans compare the perceived information about the

real-world state to their desired state, and then try to make decisions to move the real state

towards their expected state. The second loop indicates that humans make decisions not only

based on their perceptions of the real world but also based on decision rules. These decision

rules in decision-making are governed by mental models, which can be altered by information

13
from the real world. The third feedback process links ‘information feedback’ and ‘mental

models’, representing that information can change the mental models of humans, and mental

models can affect the information perceived from the real world. This model indicates that the

information feedback humans perceive from the complex real world will not only alter their

decisions in the context of existing decision strategies, but also alter their mental models

(Sterman 1994, 2000). Mental models are the internal representations of external surrounding

attributes and explain perception, reasoning and decision-making in human cognition

(Johnson-Laird 1983; Sterman 2000). They can guide human perceptions and help shape

behaviours.

We adapted the double-loop learning model (Sterman 1994, 2000) for two reasons. First,

this model includes the core of system dynamics and the basis for all decision-making in the

dynamic world (Forrester 1961), that is, the cognitive process of decision-making in mental

models. Thus, this model provides a base framework for categorising the human factor causes

of the bullwhip effect from the perspective of behaviour operations (Bendoly, Donohue, &

Schultz 2006; Bendoly et al. 2010; Gino & Pisano 2008; Loch & Wu 2005). Second, using

this model enables identifying the state-of-the-art of the behavioural causes of the bullwhip

effect from the perspective of the decision-making and learning feedback process in the

dynamic environment. This identification is particularly important because limited cognition

and information (Sterman 2000) cause non-optimal decisions; thus, the factors that can cause

these limitations in optimal decision-making must be identified. This model facilitates

identification of not only the factors that have been studied but also the factors that are

unexplored.

14
 Figure 2: Double-loop learning (Sterman 1994, 2000)

4. Aggregate Results

In this section, we present the primary results from the analysis of the final article list. We

discuss the aggregate results of the selected articles from the perspectives of the publication

year, the journal title and quality counts, the research area cluster, the methods used and the

research analysis level.

15
4.1 Publication Year and Journal Title

The results on analysing the publication year of the selected studies indicated that an

increasing number of scholars are devoting attention to the effect of human factors on the

bullwhip effect. The number of selected studies on the behavioural factors regarding the

bullwhip effect has gradually increased since 2002 and started to decline from 2015 (see

Appendix A).

The final literature pool of 53 articles was derived from 32 journals. The journal title

classification is presented in Appendix B, according to the Academic Journal Guide

(Chartered Association of Business Schools [ABS] 2018) and the SCImago Journal and

Country Rank. This classification illustrates the name of the journal and the location of the

research field. The results indicated that 23 journals were listed in the ABS and nine were not.

Most of the articles were from International Journal of Production Economics (six) and

Production and Operations Management (five), which specialise in operations and

technology management.

4.2 Major Research Cluster

We adopted a word cloud to construct and analyse the main research area clusters based on

the frequency of the keywords used in the literature pool of the 53 articles (Figure 3). This

word cloud showed that these keywords were indicative of the major research areas addressed

in the selected articles, and all keywords were relevant to behavioural factors and the bullwhip

effect. The most common keywords used included ‘bullwhip effect’, ‘supply chain

management’, ‘behavioural operations’ and ‘beer game’. Among these, ‘bullwhip effect’

appeared 25 times, ‘supply chain management’ 19 times, ‘beer game’ 11 times and

‘behavioural operations’ 10 times. Other commonly used keywords that appeared fewer than

16
10 times were ‘inventory management’, ‘beer distribution game’, ‘supply chain’, ‘simulation’,

‘system dynamics’, ‘decision making’, ‘experimental economics’, ‘dynamic decision

making’, ‘experiments’, ‘information sharing’, ‘human experiments’, ‘decision biases’,

‘supply chain coordination’, ‘inventory’, ‘inventory control’ and ‘structural equation

modelling’. Overall, this analysis indicated that the most attention has been devoted to the

areas of the bullwhip effect and supply chain management, followed by the areas of resolving

the bullwhip effect by using the beer game setting and from the perspective of behavioural

operations. Moreover, other methods and concerns of the bullwhip effect studies were

suggested by the appearance of other keywords, such as ‘decision biases’ and ‘simulation’.

Figure 3: Word cloud of most frequent keywords in reviewed articles

4.3 Analysis of Research Method and Research Level

We classified the methods used in the selected articles based on Bendoly et al.’s (2010)

summary of the typical research methods used in the study of system dynamics, group

dynamics, cognitive psychology and social psychology. The classification included case

study, controlled experiment, survey/questionnaire, modelling and simulation. Case studies

17
involve observations of human behaviours or direct communication with people in a real

working environment to explore a real-world phenomenon (e.g. Niranjan, Wagner, & Bode

2011). Controlled experiments refer to research involving manipulation by researchers to

study different treatments on different participants and groups (e.g. Croson & Donohue 2003).

In a complex environment, such as a supply chain, controlled experiments allow researchers

to control and design the environment and identify causality between behavioural factors and

empirical regularities (Donohue & Croson 2002). Surveys/questionnaires are a method

through which researchers collect reliable data from humans, by measuring the problems

under study (e.g. Moon & Kim 2005). Modelling involves establishing models to generate

accurate statements about behavioural processes to explain, predict or control human

behaviour in the laboratory or real business world (e.g. Bruccoleri, Cannella, & La Porta

2014; Mazur 2006). The computer simulation method involves studying a system that exists

or is designed by simulating a real scenario or system, and it examines the dynamic

interaction between human behaviour and complex systems (e.g. Rong, Shen, & Snyder

2008). In addition to these research methods, we considered ‘simulation experiment’ as a

research method. Unlike controlled experiments and computer simulations, this method

simulates supply chain scenarios involving subjects without controlling independent variables

or having control and experimental groups (Niranjan, Metri, & Aggarwal 2009; Di Mauro et

al. 2020), such as the ‘beer game’ introduced by Forrester (1958).

The results reflected that most studies (66%) built their simulations, models and

experiments based on the ‘beer game’ (see Appendix C for details). Thus, this research

method has been the most widely used in studies on the bullwhip effect in supply chains.

However, not all the studies used the beer game simulation in their research method. For

example, Delhoum and Scholz-Reiter (2009) used the ‘supply net game’, which is a

18
simulation game of production in networks. Among all the methods (see Figure 4), the

controlled experiment was the most widely adopted method in this sample (56.6%). The next

most common method was modelling (26.4%), followed by computer simulations (20.8%),

surveys/questionnaires and simulation experiments (17%) and, last, case studies (3.8%).

Moreover, 39.6% of the selected studies applied more than one method.

56.6%

26.4%
20.8%
17.0% 17.0%

3.8%

Figure 4: Research method applied in the literature pool

The research level depended on the behavioural content of the selected articles and the

perspective of studying this behavioural content. The research content relates to group

attributes and individual attributes, which could indicate how individual factors and social

interaction factors affect operational performance and how such attributes relate to cultural

(Gino & Pisano 2008; Loch & Wu 2005). Thus, we classified the research into three levels:

individual; group and organisational (social); and cultural. In individual-level analyses, the

behavioural content focused on the individual cognitive perspective, such as decision-makers’

thinking, perceptions, reasoning and emotions related to supply chains (Gino & Pisano 2008).

The group- and organisational-level analyses investigated the social interactions among all

19
members in the supply chain, such as trust and communication (Gino & Pisano 2008). The

cultural-level analyses studied likely differences in supply chain performance under certain

cultural backgrounds. Overall, 77.4% studies selected were undertaken at the individual level,

which means that these examined research content at the individual level of each role or each

decision-maker (e.g. Cantor & Macdonald 2009). Of this 77.4%, in all, 31.7% studied

individual cognition at the social level and 2.4% at the cultural level. The second most

common level was the social (group/organisational) level, with 47.2% of studies focused on

interaction among other members (e.g. Wu & Katok 2006). Moreover, 8% studied social

interaction at the cultural level. The least common level was the cultural level, with 3.8% of

studies undertaken at this level and based on culture (e.g. Cao, Baker, & Schniederjans 2014).

4.4 Professional v. Student Participants

The use of professional participants or student participants in beer game is arguable in the

research of bullwhip effect. Among 39 studies that applied a controlled experiment or a

simulation experiment involving human participants, 35 studies comprised student

participants, such as undergraduate, graduate and MBA students. The samples had either one

type of students or a combination of two or all types. Further, these 35 studies can be

classified into studies that included student participants with (10; e.g. MBA students) and

without work experience (25; e.g. undergraduate students). In addition, the samples of nine

studies included professional participants, among which three had only professional

participants.

Table 2 summarises the studies involving participants with work experience and indicates

whether these studies compared the results on participants when including more than one

participant type. The results illustrated that seven articles made such comparisons of their

20
experiment results, of which five studies indicated that the performance of students without

work experience was consistent with that of professionals (Croson & Donohue 2006; Sterman

1988, 1989a; Tokar et al. 2016; Turner et al. 2020) and one found no significant difference

between students with (i.e. MBA students) and without work experience (Tokar et al. 2016).

However, Tokar, Aloysius and Waller (2012) and Ancarani, Di Mauro and D’Urso (2016)

have observed that managers tend to perform better than student participants, since students

may not have better understanding of concepts than professionals and they tend to frequently

overestimate orders in the experiment.

Overall, most studies highlighted that professional participants and students do not differ

significantly in terms of their performance in the beer game experiment. However, the

participants in most studies had never played the beer game before, while few studies

repeatedly ran the beer game for the same group of students and professionals to compare

their performance. Thus, the similar performance found between both samples may be

because both were not familiar with the game rules and may have intuitively and randomly

placed orders. Consequently, we argue that professionals and MBA students with rich

working experience may perform differently from students without working experience,

driven by some behavioural factors, such as overconfidence and emotion (Ancarani, Di

Mauro, & D’Urso 2016; Sterman 1989a), although this is an area for future studies.

Participants

Authors (year) Students without Students with

work experience work experience

Professionals Comparison
(e.g. undergraduate) (e.g. MBA)

Sterman (1988) √ √ √ √

21
Sterman (1989a) √ √ √ √

Steckel, Gupta and √

Banerji (2004)

Croson and Donohue √

(2005)

Croson and Donohue √ √ √

(2006)

K-T Hung and Ryu √

(2008)

Niranjan, Metri and √

Aggarwal (2009)

Haines, Hough and √

Haines (2010)

Niranjan, Wagner √

and Bode (2011)

Tokar, Aloysius and √ √ √

Waller (2012)

Narayanan and √ √ No comparison

Moritz (2015)

Tokar et al. (2016) √ √ √ √

Ancarani, Di Mauro √ √ √

and D’Urso (2016)

Cannella et al. (2019) √

22
 Di Mauro et al. √

(2020)

Turner et al. (2020) √ √ √ √

Table 2: Summary of studies involving participants with work experience

5. Content Analysis

We modified the double-loop learning model based on the theory of behavioural operations

and the findings from the selected articles shown in Figure 5. A notable point is that we used

our modified model to represent decision-making in a single echelon in supply chains

involving social interaction with other individuals. We added three important behavioural

factors to the original model: social interaction, individual traits and emotions. Social

interaction represents the interactions between decision-makers in different echelons in a

supply chain system, regarding, for example, information sharing and coordination. Since a

supply chain is a social network consisting of many individual echelons, it is necessary to

understand how the presence of other echelons influence the current state of the real world,

and thus influence the decisions and behaviours of those in individual echelons, and the whole

supply chain network. In addition, individual traits, such as personality, age, gender and

cultural background, can affect cognitive processes (in Figure 5, the box within blue dashes

represents the factors involved in the cognitive process). We also added the factor emotion to

our modified model since it is essential in decision-making and in completing human

rationality (Simon 1983). Several studies from our literature pool indicated that participants

felt frustrated and helpless during the experiment. These emotions could be triggered in case

of a mismatch between the perceived outcomes and participants’ expectations about their last

decision (Sterman 1989a). These emotions could affect social interactions since people may

23
blame and mistrust others owing to such unexpected outcomes (Kovacevic et al. 2013;

Sterman 1989a). Moreover, emotions could influence their decision strategy and behaviours

(Nienhaus, Ziegenbein, & Schoensleben 2006; Sterman and Dogan 2015). Likewise, studies

have demonstrated that emotions could be influenced by individual traits; for instance, those

with a neurotic personality more likely tend to be anxious (Thompson 2008). However, we

found it difficult to ascertain the type of relationships between emotion and other behavioural

factors based on the information from the selected articles. Therefore, we used red dashes in

our modified model to indicate that there is a complex relationship between emotion and

social interaction, individual traits and cognitive processes (including other behavioural

factors in the box within blue dashes).

Through our modified Sterman’s (1994, 2000) framework, we classified the content of all

collected articles into seven major categories. The category of social interaction had the

subcategories of communication and coordination, information sharing/exchange and trust.

This category indicated whether the study involved interactions with other individuals that

could alter the state of the environment. The other six major categories relating to cognitive

psychology were emotion, individual traits, information feedback, mental models, decision

strategy and decisions. In addition, the modified framework shows that each main category

had subcategories that covered the content related to behavioural factors in the literature pool

of the 53 articles selected (see Figure 5).

24
Figure 5: Advanced decision-making model (adapted from Sterman 1994, 2000)

25
5.1 Social Interaction

5.1.1 Communication and coordination

According to Sterman (1994, 2000), mental models of decision-making can be altered by

social interactions with the real world. Communication and coordination are social

interactions that can provide additional information feedback about the dynamic system in

supply chains to decision-makers. These interactions can solve the bullwhip effect problem

caused by irrational decisions made with limited information. Among the selected articles,

Von Lanzenauer and Pilz-Glombik (2002), Kovacevic et al. (2013) and Wu and Katok (2006)

studied the role of these social interactions in supply chain management. By applying the

approach of contrasting model-based and human decision-making, Von Lanzenauer and Pilz-

Glombik (2002) demonstrated that centralised decision-making can produce better

performance than decentralised decision-making owing to the coordination among supply

chain members. Kovacevic et al. (2013) proposed a theoretical approach to identify the

patterns of communication through transactional analysis. An appropriate communication

pattern can result in efficient coordination, which can help reduce the bullwhip effect in a

supply chain. Wu and Katok (2006) found through the beer game experiment that the

bullwhip effect still occurs in the absence of communication and coordination even when

participants are trained with relevant knowledge; however, decision-making with

communication and coordination could improve supply chain performance.

5.1.2 Information sharing and exchange

Another form of social interaction is information sharing and exchange. By controlling all

the operational causes proposed by Lee, Padmanabhan and Whang (1997) in the beer game,

the results showed that sharing point-of-sale (POS) information could help reduce the

26
bullwhip effect in a stationary and known demand setting, and downstream inventory

information sharing could benefit upstream members and reduce the bullwhip effect, but that

sharing upstream information has no significant effect (Croson & Donohue 2003, 2005,

2006). Steckel, Gupta and Banerji (2004) examined the effect of reduced order and delivery

cycles and shared POS information on the bullwhip effect. Jin (2015) conducted a controlled

experiment with the beer game and noted that information sharing greatly reduced the

bullwhip effect. Thompson and Badizadegan (2015) also demonstrated the importance of

information sharing in reducing the bullwhip effect through an integrated analytical approach

using the beer game. Through a case study, W-H Hung, Lin and Ho (2014) highlighted that

information sharing could be affected by the degree of trust between partners; thus,

strengthening trust could help increase information sharing and reduce the uncertainties that

cause the bullwhip effect. Pamulety and Pillai (2016) studied customer demand information

sharing and indicated that such demand sharing outperforms other forms of information

sharing for reducing the magnitude of order variance. Sarkar and Kumar (2015) investigated

information sharing using the beer game in a controlled experiment and suggested that

sharing disruption information is beneficial in improving performance. Irrational decisions are

made not only because of limited information but also because of information feedback delay;

in this regard, Machuca and Barajas (2004) illustrated how to use an electronic data

interchange to reduce information feedback delay to reduce the bullwhip effect and improve

supply chain performance. However, even when information is shared with decision-makers,

the bullwhip effect still exists.

5.1.3 Trust

By increasing confidence among members in supply chains, trust could lessen uncertainties

and increase information sharing (Hung, W-H, Lin, & Ho 2014). Based on the survey method,

27
Cao, Baker and Schniederjans (2014) proposed that guanxi—as an important element of

social capital in Chinese culture and society—has a positive influence on the bullwhip effect,

and this influence would be weakened in an environment of uncertainty because of the lack of

trust. Croson et al. (2014) conducted a study on trust by eliminating all the demand variability

and forecasting without time limits in a controlled beer game experiment. Their study differed

from the other controlled experimental beer games in that it was one of the three studies

involving both participants and computer agents. Unlike the other two studies (Cantor &

Macdonald 2009; Haines, Hough, & Haines 2017), they compared the condition in which

participants manipulated all positions with that in which participants manipulated only one of

the positions and computer agents managed the remaining positions. Each participant was

informed about the presence of other participants and the computer agents and was also

provided the publicly information about an optimal decision rule. The results indicated that

treatment with all humans manipulated all positions performed worse than treatment

involving computer agents. They found that the lack of trust and the lack of knowledge about

the decision rules of the system were not only the drivers of coordination risk, but also the

behavioural causes of the bullwhip effect.

In summary, the controlled experiment is the most frequently used research method, and

the common conclusion is that building a good buyer–supplier relationship is central to

reducing the bullwhip effect. For example, if participants believe in their partners and their

abilities, it improves the supply chain performance (Croson et al. 2014). This finding is

consistent with those about numerous real-world supply chain scenarios, such as the classic

Barilla SpA (A) example (Hammond 1994). In this case, Barilla experienced the bullwhip

effect caused by the lack of coordination, information sharing and trust among suppliers.

Customers downstream focused on different objectives, without considering the effect on

28
suppliers upstream. For instance, distributors purchased more dry products during a

promotion so they could reduce purchasing costs; however, they did not consider that this

could cause demand fluctuations for their suppliers. Order fluctuations were generated by this

hoarding behaviour because of some customers’ distrust that their suppliers could not satisfy

their demand. Moreover, some of Barilla’s customers were unwilling to share POS

information, which meant that Barilla could not forecast with accuracy, even though many

articles have demonstrated such forecasting to be beneficial for the whole supply chain (e.g.

Croson & Donohue 2003). If the supply chain members are willing to cooperate and build a

strong relationship based on mutual trust, they can improve their own performance and the

performance of the entire supply chain.

5.2 Information Feedback

5.2.1 Perception

Perception is the way that information is organised, selected and interpreted, so that the

environment can be represented and understood. Perception can be shaped by other cognitive

processes, such as attention (Schacter, Gilbert, & Wegner 2011), and can be guided by mental

models. Haines, Hough and Haines (2017) investigated the effect of perception on the

decision-making process in supply chains. In a laboratory experiment of supply chain

simulation, they suggested that decision-makers in supply chains do not use all the available

information in their decision-making with decision rules. However, they will select useful

information and incorporate it into decisions based on their perception if they understand the

cause-and-effect relationship regarding the information feedback in their mental models.

5.2.2 Ambiguity

29
 Ambiguity occurs owing to a lack of information and understanding about the cause-and-

effect relationship regarding perceived information feedback (Sterman 1994, 2000). Ancarani,

Di Mauro and D’Urso (2013) introduced the concepts of demand and lead time uncertainty in

the controlled beer game setting and found that participants held fewer inventories when

higher uncertainty existed. In addition, order variance tended to be higher under stochastic

lead time in the experiment. This result can be explained by our adapted model. As a result of

limited information, decision-makers do not understand the cause-and-effect relationship

between their decisions and the effects of these decisions on the supply chain. Ambiguity

results from limited information; yet decision-makers do not understand the effects of

decisions because of the system complexity. Thus, they tend to make decisions based on

known information and causality.

5.2.3 Debiasing

Debiasing is an intervention and a method to improve decision-making in supply chains.

Based on an experiment in a single-echelon setting, Tokar, Aloysius and Waller (2012) tested

the effects of debiasing on the inventory level, the order decision-making and the bullwhip

effect and found that training on procedural knowledge can help improve the performance of

supply chains. We categorised debiasing under information feedback because the results of

applying knowledge gained through such training are perceived as information feedback to

decision-makers. With feedback on information obtained through training, decision-makers

are able to understand the cause-and-effect relationships and can thus alter their mental

models and form new decision strategies.

The information feedback and mental models are highly related in the double-loop learning

model. Perceived information feedback could help to change the mental model or to

30
determine whether the feedback is consistent with the expectations in the existing mental

model. Moreover, the information feedback is perceived based on the existing mental model

about cause-and-effect relationships, and then, appropriate decision rules are adapted to make

decisions. This category includes only three articles, which discussed how perceived

information feedback influences the decision-making process. The primary contribution of

these articles was that decision-makers could make better decisions on understanding the

cause-and-effect relationships among perceived feedback. Training and learning could help

them better understand the causality among information feedback. Then, the mental model

could be altered and could help decision-makers to seek useful information based on the

feedback about causality. The difference between these three studies is that they used

different information and supply chain settings. Haines, Hough and Haines (2017)

manipulated information on the supply line and consumer demand, whereas Ancarani, Di

Mauro and D’Urso (2013) manipulated information on lead time and demand, but not the

actual demand or inventory sharing. Last, Tokar, Aloysius and Waller (2012) considered

complete information sharing and found that too much information had a negative effect on

performance. They tested both single-echelon and four-echelon settings, and the other two

studies investigated only four-echelon settings. We suggest that more studies be extended to

include other research settings, rather than only four echelons, such as through case studies

with a single-echelon setting, for an in-depth study of the whole supply chain network, thus

addressing the limitations of the beer game.

5.3 Mental Models

5.3.1 Rationality

31
 Rationality refers to reasoning to achieve goals within mental models. However, cognitive

limitations and imperfect mental models cause bounded rationality and affect decision-

making (Simon 1947; Sterman 2000). Some selected articles study rationality from different

perspectives.

Cantor and Katok (2012), Riddalls and Bennett (2003) and Su (2008) studied bounded

rationality. By using a two-echelon supply chain setting, Cantor and Katok (2012) smoothed

production when customer demand was seasonal; the smoothing behaviour became more

significant when the cost of changing an order was high. Riddalls and Bennett (2003)

redefined the beer distribution game as a time delay controlling system and demonstrated that

instability can be improved once the bounded rationality level in the system is overcome. To

further explore bounded rationality and the bullwhip effect, Su (2008) applied the quantal

choice model to simulate bounded rationality and indicated that bounded rationality can

explain the bullwhip effect, since humans do not always make optimal decisions. These

studies applied modelling and simulation to connect the experiment and theories, to stimulate

future behavioural studies regarding the bullwhip effect.

Procedural rationality refers to collecting information to make rational decisions in the

process of pursing goals (Dean & Sharfman 1993). Using a controlled experiment, Haines,

Hough and Haines (2010) studied procedural rationality and found that decision-makers made

their choice depending on the analysis of their perceived information feedback from the

experiment. Using different treatments of information availability, they found that

performance was most affected by the procedural rationality of decision-makers in the retailer

role and that providing limited information can cause the bullwhip effect.

32
 Cantor and Macdonald (2009) studied the problem-solving approach. They investigated the

bullwhip effect under different degrees of information availability, using a 2 × 2 experimental

design of the beer game framework. They found that under limited information sharing,

individuals with concrete (low-level) problem-solving skills perform worse than do those with

abstract (high-level) problem-solving skills. In addition, having more system information does

not result in better decisions and performance. This result may be attributable to the fact that

people do not have the mental model to perceive and understand the causality between

information and decisions; thus, they do not have effective feedback to alter their current

mental models and decision rules for decision optimisation.

Moon and Kim (2005) tested how the bullwhip effect changed based on the relationship

between consistency and individual systems thinking ability. This ability refers to an

individual’s ability to solve a problem in a complex environment or system. Based on the

survey, test and simulation research methods, the results indicated that individuals with high

consistency and high systems thinking ability produce less bullwhip effect. This study

demonstrated that a broad, holistic view of the dynamic system will have a positive influence

on supply chain management. Moreover, determining how to develop systems thinking

abilities should be the rationale underlying studies on human factors in supply chains.

Cognitive reflection is another behavioural factor under the rationality category.

Narayanan and Moritz (2015) stated that cognitive reflection is a structured analytical

problem-solving process that enables individuals to overturn or approve an immediately

available answer. They argued that making decisions about supply line underweighting is

related to the level of individual cognitive reflection, and that individuals with higher

cognitive reflection have higher performance. The incorporation of perceived information in

the decision-making process plays a key role in enabling the understanding of the causality of

33
known information during problem-solving and analysis. Thus, future studies should examine

how to improve relevant abilities in problem-solving to enable decision-makers to seek

information feedback and alter mental models in a dynamic environment.

Sterman (1994, 2000) revealed that the misperception of feedback influences decision-

makers’ ability to understand the structure of the dynamic world from information feedback.

Using a simulation experimental method, Sterman (1989a, 1989b) identified misperception of

feedback during the decision-making process and considered it as the driver of players’ poor

dynamic performance. One type of misperceived feedback was misperception of time delay,

in which participants tended to ignore the delay between a control action and its full effect

and tended to perform correction behaviour in capital stock. The second type was

misperception of feedback between the environment and participants’ decisions. This

indicated that participants misunderstood information about the dynamic system when they

placed orders or did not understand the dynamic system at all. As a result of their

misperception of feedback from the environment and decisions, decision-makers cannot form

accurate mental models and optimal decisions. The misperception of feedback view was

strongly supported by Diehl and Sterman (1995) and Bloomfield and Kulp (2013). The former

conducted an experiment with stochastic sales and indicated that feedback misperception

worsened as the cause and effect increased. The latter found that the durability of inventory

and transit delays decreased optimal decisions, even within one single-echelon experimental

setting.

5.3.2 Judgement

The anchoring and adjustment heuristic can be considered as judgemental bias, which can

affect decision-makers’ mental model. Sterman (1988) stated that a simple heuristic method

34
in judgement and behavioural decision theory could explain the behaviour of participants.

Using modelling and a simulation experiment, he demonstrated that the heuristic that

participants use can produce chaos in decision-making processes. This study created the

foundation for studies on human factors of the bullwhip effect. By modelling Sterman’s

(1989a) experiment, Macdonald, Frommer and Karaesmen (2013) simulated the beer game

and proposed that long-term performance cannot be forecast by short-term performance, and

strong underweighting will not only cause the bullwhip effect but also affect the period of

chaos before the system reaches final stability. Villa, Gonçalves and Arango (2015) tested

anchoring and adjustment, similar to Sterman (1989a), with a linear econometric model, and

suggested that participants were unable to process the effects of delay and information

feedback. Through systematic dynamic modelling of human judgemental interventions,

Syntetos et al. (2011) explained the sources of the bullwhip effect in customer–supplier

chains. They stated that human judgemental interventions strongly influence adjustments in

forecasting and ordering. Zhao and Zhao (2015) decreased the biases of anchoring and

underweighting with linear regression models, with full information shared during the

experiment. Li and Yan (2015) found that different behavioural adjustment levels can lead to

different performance levels because of individual differences, and that adjustment behaviour

has less effect on performance under demand uncertainty than under supply uncertainty. They

also suggested that over-adjustment is a behaviour that should be avoided in the decision-

making process.

Further, Tokar et al. (2016) studied the framing effect bias through conducting

experimental studies. This bias refers to making decisions based on decision-making choices

framed as gains or losses (Tokar et al. 2016). In their study, negative frame resulted in greater

35
inventory and orders; understocking occurred when orders and inventory in the positive

frame, which can be debiased by framing decisions as losses.

Overconfidence has been defined as a judgemental bias in mental models that affects

reasoning in decision-making (Sterman 1994, 2000). In most economic activities, people tend

to be overconfident or overly optimistic about their knowledge, their ability level and the

accuracy of the information they receive, and hence tend to underestimate uncertainties (Dessí

& Zhao 2018). Ancarani, Di Mauro and D’Urso (2016) used the beer game to investigate

overconfidence in supply chain ordering decisions and found that overconfidence may lead to

the bullwhip effect and that environmental uncertainty is a cause of overconfidence. In

contrast to earlier studies, their participants were familiar with supply chain management.

This fact led us to understand that in the experimental environment, even experts can be

affected by overconfidence in their decisions. Shee and Kaswi (2016) also found that the

managers from a local and multinational supermarket in Indonesia were overconfident about

their decisions, which caused increasing variability to create the bullwhip effect. Unlike

Ancarani, Di Mauro and D’Urso (2016), they used surveys to investigate a real situation,

which helped them realise the true problem. However, the study only tested one country; thus,

it was difficult to determine whether there would be different results in other countries.

Supply line underestimation/underweighting has been studied in many articles. Sterman

(1989a) modelled decision-makers’ behaviour in stock management and found that

participants tended to ignore the supply line and underestimate the time delay between an

action and its effect. His study influenced later research on behaviour theories for the

bullwhip effect. Based on his study, Croson and Donohue (2003, 2006) also demonstrated that

supply line underweighting contributed to the bullwhip effect, even though their controlled

experiments allowed information sharing. To further test the results from prior studies,

36
Niranjan, Metri and Aggarwal (2009) re-examined the study methods and found that subjects

not only ignored or underweighted the supply line but sometimes even distributed a negative

weight for the supply line. Moreover, Niranjan, Wagner and Bode (2011) presented a

‘correction model’ based on a simple recurrent network and stated that over-ordering occurs

because of considering the supply line, rather than ignoring the supply line. By developing a

system dynamics model, Udenio, Fransoo and Peels (2015) suggested that underweighting the

supply line was not only caused by individual decision-making biases but also by the

combination of organisations’ reaction speed and a decision rule, and this decision rule

prevents decision-makers from tracking the supply line by controlling the number of on-hand

inventories.

In contrast, few studies have focused on the overestimation/overweighting of the supply

line. Using simulation, Rong, Shen and Snyder (2008) focused on uncertainties from both the

supply and demand sides and demonstrated that supply line overweighting and overreaction to

shocks of capacity were the behavioural reasons for the reverse bullwhip effect in supply

chain management. In addition, Udenio et al. (2017) used a control theoretic model to analyse

the effects of behavioural biases on the stability and performance of supply chains. They

demonstrated that underestimating the supply line can produce the bullwhip effect with

demand shocks, which cannot be limited by order smoothing. They found no differences

between the performance caused by supply line estimation biases and by unbiased policies

when demand was stationary. Under this condition, order smoothing could improve supply

chain performance, but only when demand was unpredictable in the worst case of order

amplification.

This category covered 26 articles from our literature pool, indicating that most researchers

focus on mental model–related causes of the bullwhip effect—the fundamental element of

37
studying system dynamics and building models. The most frequently used research method

was a controlled experiment (thirteen articles), followed by modelling (nine articles). By

using a controlled experiment in a specific setting, different variables can be manipulated in

treatment groups, and then compared with the control group. This enables understanding the

cause-and-effect relationship between biases and the bullwhip effect, which can help build

models and simulate optimal decisions based on this understanding in more complex

scenarios. The information resulting from modelling and simulation could help individuals

alter mental models, learn new decision rules and optimise decisions in the real world.

Moreover, the studies in this category mainly focused on decision-making biases, such as the

misperception of feedback, the anchoring and adjustment heuristic, overconfidence and

underweighting/overweighting the supply line. This indicates that with limited information

and limited understanding of the information based on current mental models, human tend to

use heuristics to make decisions. These decisions could also be influenced by other biases.

However, based on the findings, these decisions could be improved by obtaining more

information and more knowledge about this information.

38
5.4 Decision Strategy

5.4.1 Decision patterns

By implementing the supply net game, a simulation game of production networks, which

differs from the beer game, Delhoum and Scholz-Reiter (2009) applied and assessed six

different decision patterns of game participants: immoderate intuitive, selective intuitive,

moderate intuitive, counterintuitive, collaborative and deceptive routines. Among these

patterns, the counterintuitive pattern was the cause of the bullwhip effect since the players in

this pattern continued to place orders even when inventories were full. The collaborative

decision-making pattern allowed the participants to integrate production plans and design a

more balanced production network through collaboration (without demand information).

5.4.2 Risk aversion

We placed risk aversion in this main category because it is a strategy or decision rule that

decision-makers apply under risk. K-T Hung and Ryu (2008) found that changing the risk

preferences of participants for demand variance was a critical behavioural factor that

explained the deviation in ordering decisions in the bullwhip effect. Through controlled

experiments and modelling, they found that order quantities were based on participants’

expectations, as the degree of risk seeking altered with a change in supplier shortage and

demand deviation. When the outcomes matched participants’ expectations, they preferred to

be risk averse. Contrarily, if the outcomes mismatched their expectations, they preferred to be

risk seeking. Cannella et al. (2019) applied the method of human experiment and multi-agent

simulation to study risk aversion in a setting of a multi-echelon supply chain. To reduce the

risk of stock out, inventory holding behaviour was observed in risk-averse participants when

making order decisions. Di Mauro et al. (2020) studied the effect of risk aversion on

39
replenishment decisions in the context of a multi-echelon supply chain. The results indicated

that people with high risk aversion tended to display hoarding behaviour and could be less

affected by experiential learning.

Thus, four articles fell into this category. All studies were undertaken with a multi-echelon

supply chain. One study used a simulation experiment, rather than the beer game, to study

decision patterns. The other three focused on risk aversion, with two of them observing

hoarding behaviour when no information sharing and no communication were allowed, and

one study proposing that information sharing could change risk aversion to risk seeking.

5.5 Decisions

Decisions can be considered the behavioural implementation after the decision-making

process. Several studies have focused on the behaviour caused by decision-making. Through a

simulation experiment of the beer game, Nienhaus, Ziegenbein and Schoensleben (2006)

studied hoarding behaviour and panic reactions. Using a sample of more than 4,000

participants, they found that these two behaviours caused the underestimation of information,

which amplified the bullwhip effect. However, the demand information in their experiment

was not fully known to the participants, although this was supplemented in a later study on

irrational behaviour. Sterman and Dogan (2015) observed hoarding behaviour in the context

of the beer game through modelling and illustrated that hoarding and phantom behaviours

were more likely to be triggered by real-world stressors. For instance, poor supplier delivery

performance and increasing demand can be the stressors that trigger irrational decisions and

behaviours. Real-world stressors also lead to inaccurate behaviour in inventory recording

(Bruccoleri, Cannella, & La Porta 2014). Bruccoleri, Cannella and La Porta (2014) proposed

40
that stress (e.g. workload) can lead workers to make mistakes because of psychological

sensitivity. This inaccuracy could also contribute to the bullwhip effect.

In all, three articles examined the actual behaviours of humans. The findings illustrated

that environmental stressors and fear of having too much or too little inventory can influence

rational decision-making. Few settings and methods have been used to study external factors,

such as stressors, that may influence human behaviour when implementing decisions.

5.6 Individual Traits

5.6.1 Personality

Personality is an influential factor in decision-making processes and varies from individual

to individual (Byrne, Silasi-Mansat, & Worthy 2015). Khan, Ahmed and Hussain (2019)

analysed the bullwhip effect using the personality trait approach through the beer game and a

questionnaire and noted a negative relationship between conscientiousness and the bullwhip

effect. Bloomfield and Kulp (2013) examined the variables of the need for cognition,

impulsiveness and the locus of control related to personality through a questionnaire and

found that inventory durability and transit delay may be altered by these personality variables.

5.6.2 Age and experience

Recently, Turner et al. (2020) focused on the influence of age and experience on supply chain

performance through the beer game simulation experiment. They compared the results of

managers with those of students. The managers were older and had at least 15 years of

professional experience in production agriculture, and the students were mostly younger

undergraduates without managerial experience. In this case, the age of participants was

proportional to their professional work experience. This study revealed that more experienced

41
managers did not perform better than less experienced students and that students tended to be

more heavily anchored to the inventory level resulting in lower cost.

Thus, the studies under this category mainly focused on individual differences in decision-

making for inventory management. One mainly focused on personality, and the other

discussed findings about personality based on a survey. The results of the last article provide a

basis for the fact that managers do not perform much better than students. In addition, we

suggest that more research should be undertaken in this category within different supply chain

settings, such as different echelons, different levels of information sharing and different

disruptions, to find the impact of different individual traits on system dynamics.

5.7 Emotions

Emotion-induced irrational behaviours and decisions are present not only in experimental

studies but also in the real world. Two noteworthy points are identified on comparing the

emotional influence on participants in a laboratory experiment and in real life. First,

participants in both settings could panic and feel frustrated on not getting what they want

(Croson & Donohue 2005, Kovacevic et al. 2013; Pamulety & Pillai 2016; Sterman 1989a).

This could result in panic buying and hoarding behaviour by participants in experiments and

in real life when they are afraid that suppliers cannot satisfy their demand and begin to

increase their orders and stock (Nienhaus, Ziegenbein, & Schoensleben 2006; Sterman &

Dogan 2015). Participants tend to use this behaviour and strategy to cope with the fear and

anxiety caused by the unknown. Second, it is difficult for the laboratory experiment setting to

replicate some scenarios, such as the supply chain disruptions caused by natural disasters or

human-made disasters in real life. Therefore, the emotions of participants in the laboratory

may be less intense than those of real-life participants who face real supply chain disruptions

42
and unexpected risks. For instance, a toilet paper shortage has emerged in Western countries

because of the coronavirus disease pandemic. Nobody anticipated the sudden onset and the

severity of this pandemic. People feel anxious and helpless when they see the empty shelves

in stores and they do not know what to do. Therefore, under the influence of such emotions,

when they next see toilet paper in stock, they will buy more than they actually need and try to

stock as much as they can to cope with their anxiety and fear of the future stockout of toilet

paper. In turn, this sudden increase in demand and the consequent disruptions in supply will

lead suppliers in the supply chain to feel the emotion of panic and to engage in hoarding

behaviour.

6. Conclusion and Future Research Directions

In this study, we aimed to demonstrate the importance of human behavioural factors in

analysing the bullwhip effect and to identify research gaps by considering possible problems

in real business scenarios through a systematic review. We selected and analysed 53 peer-

reviewed academic studies, using a systematic literature search procedure. The results

indicated that behavioural factors involving social interactions and cognitive processes are

important in the decision-making process in a dynamic system. Cognitive limitations and

incomplete mental models can result in bounded rationality in decision-making and

negatively influence supply chain performance (Sterman 1994, 2000).

We adapted Sterman’s (1994, 2000) double-loop learning model to classify the behavioural

factors revealed in the collected articles into seven main categories: social interaction,

information feedback, mental models, decision strategy, decisions, individual traits and

emotion. The results of each main category reflected the social and cognitive factors that have

been studied in dynamic decision-making. Among these categories, most articles based their

43
studies on the beer game experiment (66%), whereas the controlled experiment method was

the most used research method (56.6%). The category of social interaction revealed the

importance of interaction involving communication, information sharing and trust in supply

chain coordination. Further, in cognitive psychology, by applying the research methods of

surveys/questionnaires or controlled experiments, the effects of individual traits (including

personality) were demonstrated. The information feedback category demonstrated that the

perceptions of information, ambiguity and debiasing can influence the information perceived

from the real world and result in failure or success in altering mental models in decision-

making. In addition, 26 studies focused on the mental model category, and were classified

into the two main subcategories of rationality and judgement. The mental model category

involved the largest number of studies, which indicated its importance in the behavioural

factor research on the bullwhip effect. The next largest category was social interaction, which

had 15 articles. The remaining categories had fewer than six articles. Further, the decision

strategy and decision categories were shown to influence the bullwhip effect through

controlled experiments, simulation experiments and computer simulation. In addition, the

emotion category needs more research to determine its relationship with other behavioural

factors.

All these main results implied that improvements to individual mental models would help

to mitigate the bullwhip effect and to optimise supply chain decision-making. Moreover, the

aggregated results indicated that most of the studies concentrated on analysing behavioural

factors at the individual level, and few have considered behavioural factors at the social and

cultural levels. Based on our research findings, we provide the following suggestions from

various aspects regarding potential future research directions:

44
 Research-level aspect: Most of the studies were conducted at the individual level. This

finding leads us to call for the study of the behavioural causes of the bullwhip effect at

the group level. Group decisions have been demonstrated to improve the quality of

decision-making in the context of inventory management. In the newsvendor problem,

group decisions debias the anchoring and adjustment in individual decision-making

when receiving various pieces of information (Gavirneni & Xia 2009). Based on our

findings, debiasing could help improve supply chain performance and reduce the

bullwhip effect at the individual level (Tokar, Aloysius, & Waller 2012). Thus, it is

worth determining the effect of group decisions on the bullwhip effect and whether

group decisions result in better performance than individual decisions. Moreover,

further studies should be conducted at the cultural level. Decision-makers from

different cultures may have mental models that capture different information and

result in different decisions. The ‘pull-to-centre’ effect of Chinese decision-makers

tends to be more significant in the newsvendor problem than does that of American

decision-makers because of ‘the Doctrine of the Mean’ in the Chinese culture (Feng,

Keller, & Zheng 2011). No article has yet examined the bullwhip effect in terms of the

effect of group decisions and cultural differences on decision-making. Thus, this is a

research direction for future bullwhip effect studies.

 Social psychology aspect: Cultural factors (e.g. cultural differences and fairness),

which have been found to be important in making decisions, little research considered

in the selected articles (Ng, Lee, & Soutar 2007; Podrug 2011; Schramm 2001). In

addition, trust in social interactions was studied in a controlled beer game experiment

by comparing the condition in which participants manipulated all positions with the

condition in which participants completed the inventory task individually and

45
 computer agents managed other positions (Croson et al. 2014). However, it could be

extended to consider whether trust issues will decrease or increase with an increase in

the number of human participants in the experiment. Further, trust has not been

measured at different levels in different scenarios. Moreover, other factors that could

influence buyer–supplier relationships, such as commitment and contiguity (e.g. Zhao

et al. 2008), have not been studied in terms of the behavioural elements of the

bullwhip effect. Thus, further research can be conducted to fill the remaining gaps

from the social psychology aspect, including on social norms and culture that can

affect the supply chain management of multinational enterprises. In addition, the

unexplored behavioural factors in buyer–supplier relationships involving social

interactions provide avenues for future research.

 Cognitive psychology aspect: In this study, we adopted Sterman’s (1994, 2000)

double-loop learning model. This model involves the cognitive processes that

influence how humans perceive environmental elements and detect useful information

to make effective decisions (see the box within blue dashes in Figure 5). Scholars have

studied the behavioural causes of the bullwhip effect in each category involving

cognitive processes. Most studies investigated mental models, and we categorised only

three as considering information feedback; thus, this area requires further study.

Further, other factors in information feedback, such as attention (Simon 1957), have

not been studied until date. The categories of decision strategy, decisions and

individual traits require more attention, given the limited studies focusing on these two

categories. The influence of emotions was highlighted as a crucial factor causing the

bullwhip effect in selected studies. Helplessness, frustration and blaming have been

observed during decision-making (Croson & Donohue 2005; Kovacevic et al. 2013;

46
 Pamulety & Pillai 2016; Sterman 1989a), and stressors created by the environment

have been found to affect emotional state, resulting in irrational decisions and

behaviours (Nienhaus, Ziegenbein, & Schoensleben 2006; Sterman & Dogan 2015).

Irrational behaviour, such as panic buying, can result in supply chain disruptions not

only in experiments but also in real life. Thus, future studies should consider the

influence of emotions on decision-making and behaviours, and more studies should be

undertaken in the category of information feedback, decision strategy, decisions and

personality, and their relationships with emotion.

 Research method aspect: Overall, most studies adopted the beer game simulation

experiment as the research method. However, this approach has a few limitations.

Although it is able to represent the dynamic system of supply chains in a laboratory

experiment, the setting is not as complex as a real-world business scenario. In

addition, this method takes some time to introduce to the participants, and the

experimental process is relatively slow if involving human participants. Hence, future

studies should consider case studies as the research method to determine the effect of

behavioural causes in the real world, because case studies can identify and describe the

key variables and could thus represent the supply chain management field (Stuart et al.

2002). Moreover, case studies have the advantage of offering deep insights into human

behaviour within a specific context (e.g. the decision-making environment and

industrial settings). In addition, the category of social interactions and mental models

should be investigated using research methods other than the controlled experiment.

Further, more studies should compare the performance of professional and MBA

student participants with rich work experience with that of student participants without

experience to identify the differences in performance.

47
 Research setting aspect: More studies on the bullwhip effect should be undertaken in

more complex research settings that are closer to real-world scenarios, rather than

deterministic settings. Most studies applied deterministic beer game settings, and few

applied a stochastic variable setting, such as stochastic lead time and demand, which

contribute to worsening supply chain stability in the real world. In addition, studies on

the bullwhip effect should extend the laboratory setting from the traditional forward

supply chains to closed-loop supply chains based on the beer game. Further,

considering that only Sarkar and Kumar (2015) studied supply chain disruption, more

supply chain disruption settings could be introduced to the beer game to gain insight

into the performance of supply chain systems under sudden change. This could help to

better understand the effects of supply chain disruptions and to identify ways to

mitigate these effects.

 Supply chain digitalisation and automation aspect: Digital technologies, including

cloud computing, artificial intelligence, blockchain technology, autonomous robots,

cyber-physical systems and additive manufacturing, have been applied to supply

chains and may be able to reduce the bullwhip effect (Wiedenmann & Größler 2019).

For example, blockchain technology could increase the information transparency and

reliability in supply chains and thus reduce the bullwhip effect (Van Engelenburg,

Janssen, & Klievink 2018). These technologies provide a way for supply chain

networks to integrate and manage information, resources and materials, which could

help to optimise the related decision-making and realise the optimal combination of

cost, efficiency and experience. Therefore, the impact of human decision-makers on

mitigating the bullwhip effect may depend on the degree of digitalisation of supply

chains. From such perspective, bullwhip effect can be reduced if decision markers can

48
 trust more in artificial intelligence (AI) agent rather than other humans such as other

supply chain members (Croson et al. 2014). However, high level of automation may

lead to the negative impact on bullwhip as decision markers tend to over trust AI or

put less efforts and thoughts on decision marking, although the influence of human

factors on the bullwhip effect has not been explored in the supply chain digitalisation

context. Thus, more studies should include supply chain digitalisation in their research

setting or should consider the context of Industry 4.0 to investigate the impact of

incremental supply chain automation on the behavioural causes of bullwhip in the

future.

Thus, we have contributed to this research field by systematically reviewing and analysing the

behavioural causes of the bullwhip effect, synthesising existing academic findings and

indicating a future research agenda. We have demonstrated that human behaviours are a factor

that cannot be ignored in considering the bullwhip effect, since human mental models are

significant in dynamic decision-making. By studying, understanding and analysing mental

models and the constraints to improving mental models, we can gradually alter mental models

based on the enhanced understanding of cause-and-effect relationships of information

feedback. Thus, we can make better decisions and optimal decisions eventually. However,

this study has some limitations, even though we conducted our search process carefully and

strictly. We used the databases of Scopus, Science Direct and Wiley Online Library and

performed cross-checking with Google Scholar; however, different search results may be

found, given that other literature databases have different search terms. Further, considering

that different authors and disciplines use different terminologies, the keywords and

terminologies are inconsistent in the literature. Thus, we may have overlooked articles

because some studies may have used incorrect terminologies or may not have used the

49
terminologies that we used as keywords during the source identification stage. Last, we may

also have overlooked articles related to our research scope because we were unable to exhaust

all possible keywords.

50
7. Appendices

Appendix A: Linear Growth in Research Based on Publication Year

10

6
Number of Papers

0
1988 1991 1994 1997 2000 2003 2006 2009 2012 2015 2018 2021

Publication Year

51
Appendix B: Journal Classification

Research Field Journal Title Number of Reviewed Articles ABS

Articles
Operations and technology Journal of Operations Management 3 Cantor and Macdonald (2009), Sterman and Dogan 4*
management (2015), Wu and Katok (2006)
Production and Operations Management 5 Bloomfield and Kulp (2013), Croson and Donohue (2003), 4
Croson et al. (2014), Narayanan and Moritz (2015), Tokar
et al. (2016)
International Journal of Production Economics 6 Ancarani, Di Mauro and D’Urso (2013), Cao, Baker and 3
Schniederjans (2014), Haines, Hough and Haines (2017),
Sarkar and Kumar (2015), Udenio, Fransoo and Peels
(2015), Zhao and Zhao (2015)
International Journal of Production Research 1 Cannella et al. (2019) 3
Manufacturing and Service Operations Management 1 Su (2008) 3
Production Planning and Control 3 Delhoum and Scholz-Reiter (2009), K-T Hung and Ryu 3
(2008), Nienhaus, Ziegenbein and Schoensleben (2006)
Supply Chain Management: An International Journal 1 Moon and Kim (2005) 3
International Journal of Physical Distribution and 1 Bruccoleri, Cannella and La Porta (2014) 2
Logistics Management
Journal of Business Logistics 1 Haines, Hough and Haines (2010) 2
Journal of Purchasing and Supply Management 2 Ancarani, Di Mauro and D’Urso (2016), Di Mauro et al. 2
(2020)
Flexible Services and Manufacturing Journal 1 Rong, Shen and Snyder (2008) 1
International Journal of Logistics Research and 1 W-H Hung, Lin and Ho (2014) 1
Application
International Journal of Services and Operations 1 Niranjan, Metri and Aggarwal (2009) 1
Management
Operations Management Research: Advancing Practice 1 Macdonald, Frommer and Karaesmen (2013) 1
through Theory
Supply Chain Forum: An International Journal 1 Khan, Ahmed, S & Hussain (2019) 1

52
Operations research and Management Science 3 Croson and Donohue (2006), Steckel, Gupta and Banerji 4*
management science (2004), Sterman (1989a)
Decision Science 2 Niranjan, Wagner and Bode (2011), Tokar, Aloysius and 3
Waller (2012)
IISE Transactions (IIE Transactions) 1 Udenio et al. (2017) 3
Journal of the Operational Research Society 1 Syntetos et al. (2011) 3
OR Spectrum 1 Von Lanzenauer and Pilz-Glombik (2002) 3
System Dynamics Review 3 Croson and Donohue (2005), Sterman (1988), Villa, 2
Gonçalves and Arango (2015)
Psychology Organizational Behavior and Human Decision 2 Sterman (1989b), Diehl and Sterman (1995) 4
(organisational) Processes
Sector studies Transportation Research Part E: Logistics and 2 Cantor and Katok (2012), Machuca and Barajas (2004) 3
Transportation Review
Economics, econometric Amfiteatru Economic 1 Kovacevic et al. (2013) Non-
and finance ABS
Multidisciplinary Complexity 1 Edali and Yasarcan (2016) Non-
ABS
Systems 1 Turner et al. (2020) Non-
ABS
Mathematics Discrete Dynamics in Nature and Society 1 Li and Yan (2015) Non-
ABS
Computer science; Information (Japan) 1 Jin (2015) Non-
multidisciplinary ABS
Computer science, International Journal of Systems Science 1 Riddalls and Bennett (2003) Non-
engineering and ABS
mathematics
Computer science IEEE Access 1 Thompson and Badizadegan (2015) Non-
ABS
Business, management Uncertain Supply Chain Management 1 Pamulety and Pillai (2016) Non-
and accounting ABS
Operations and supply Operations and Supply Chain Management: An 1 Shee and Kaswi (2016) Non-
chain management International Journal (OSCM) ABS

53
Appendix C: Research Methodology Classification

Reviewed Research Method Major Behavioural Category

Articles by Case Controlled Survey/ Modelling Simulation Computer Social Individual Information Mental Decision Decision
Year Study Experiment Questionnaire Experiment Simulation Interaction Traits Feedback Models Strategy
Sterman √ √ (BG) √
(1988)
Sterman √ √ (BG) √
(1989a)
Sterman √ √ √
(1989b)
Diehl and √ √
Sterman
(1995)
Haehling and √ √ (BG) √
Pilz-Glombik
(2002)
Croson and √ (BG) √
Donohue
(2003)
Riddalls and √ √
Bennett
(2003)
Machuca and √ (BG) √
Barajas (2004)
Steckel, Gupta √ (BG) √
and Banerji
(2004)
Croson and √ (BG) √
Donohue
(2005)
Moon and √ (BG) √ √ √
Kim (2005)

54
Croson and √ (BG) √
Donohue
(2006)

Nienhaus, √ (BG) √
Ziegenbein
and
Schoensleben
(2006)
Wu and Katok √ (BG) √ √
(2006)
Su (2008) √ √
Rong, Shen √ (BG) √ √
and Snyder
(2008)
K-T Hung and √ (BG) √ √
Ryu (2008)
Niranjan, √ (BG) √
Metri and
Aggarwal
(2009)
Cantor and √ (BG) √
Macdonald
(2009)
Delhoum and √ (SNG) √
Scholz-Reiter
(2009)
Haines, √ (BG) √ √
Hough and
Haines (2010)
Syntetos et al. √ √
(2011)
Niranjan, √ √ (BG) √
Wagner and
Bode (2011)
Cantor and √ (BG) √
Katok (2012)

55
Tokar, √ (BG) √
Aloysius and
Waller (2012)
Kovacevic et √ (BG) √ √
al. (2013)
Ancarani, Di √ (BG) √ (BG) √
Mauro and
D’Urso (2013)
Bloomfield √ √ √
and Kulp
(2013)
Macdonald, √ (BG) √
Frommer and
Karaesmen
(2013)
Bruccoleri, √ √ √
Cannella and
La Porta
(2014)
Cao, Baker √ √ √
and
Schniederjans
(2014)
W-H Hung, √ √
Lin and Ho
(2014)
Croson et al. √ (BG) √
(2014)
Sarkar and √ (BG) √
Kumar (2015)
Narayanan √ (BG) √
and Moritz
(2015)
Li and Yan √ √ √
(2015)
Villa, √ √ √

56
Gonçalves and
Arango (2015)
Zhao and √ √
Zhao (2015)
Udenio, √ √
Fransoo and
Peels (2015)
Jin (2015) √ (BG) √
Sterman and √ √
Dogan (2015)
Thompson √ (BG) √ (BG) √
and
Badizadegan
(2015)
Tokar et al. √ (BG) √
(2016)
Ancarani, Di √ (BG) √
Mauro and
D’Urso (2016)
Edali and √ (BG) √
Yasarcan
(2016)
Pamulety and √ (BG) √
Pillai (2016)
Shee and √ √
Kaswi (2016)
Haines, √ (BG) √ √
Hough and
Haines (2017)
Udenio et al. √ √
(2017)
Cannella et al. √ (BG) √ √
(2019)
Khan, Ahmed √ √ √
and Hussain

57
(2019)
Di Mauro et √ √ (BG) √
al. (2020)
Turner et al. √ (BG) √
(2020)
Total number 2 30 (25 BG) 9 14 9 (8 BG) 11 (4 BG) 15 3 3 26 4 3
Note: BG represents the ‘beer game’, and SNG represents the ‘supply net game’.

58
8. References

Ancarani, A, Di Mauro, C & D’Urso, D 2013, ‘A human experiment on inventory decisions

under supply uncertainty’, International Journal of Production Economics, vol. 142, no.1, pp.
61–73, https://doi.org/10.1016/j.ijpe.2012.09.001.
Ancarani, A, Di Mauro, C & D’Urso, D 2016, ‘Measuring overconfidence in inventory
management decisions’, Journal of Purchasing and Supply Management, vol. 22, no. 3, pp.
171–80, https://doi.org/10.1016/j.pursup.2016.05.001.
Arvan, M, Fahimnia, B, Reisi, M, & Siemsen, E 2019, ‘Integrating human judgement into
quantitative forecasting methods: a review’, Omega, vol. 86, pp. 237–52,
https://doi.org/10.1016/j.omega.2018.07.012.
Baecke, P, De Baets, S & Vanderheyden, K 2017, ‘Investigating the added value of
integrating human judgment into statistical demand forecasting systems’, International
Journal of Production Economics, vol. 191, pp. 85–96,
https://doi.org/10.1016/j.ijpe.2017.05.016.
Bendoly, E, Croson, R, Goncalves, P & Schultz, K 2010, ‘Bodies of knowledge for research
in behavioral operations’, Production and Operations Management, vol. 19, no. 4, pp. 434–
52, https://doi.org/10.1111/j.1937-5956.2009.01108.x.
Bendoly, E, Donohue, K & Schultz, KL 2006, ‘Behavior in operations management:
assessing recent findings and revisiting old assumptions’, Journal of Operations
Management, vol. 24, no. 6, pp. 737–52, https://doi.org/10.1016/j.jom.2005.10.001.
Bhattacharya, R & Bandyopadhyay, S 2011, ‘A review of the causes of bullwhip effect in a
supply chain’, International Journal of Advanced Manufacturing Technology, vol. 54, no. 9–
12, pp. 1245–61, https://doi.org/10.1007/s00170-010-2987-6.
Bloomfield, RJ & Kulp, SL 2013, ‘Durability, transit lags and optimality of inventory
management decisions’, Production and Operations Management, vol. 22, no. 4, pp. 826–42,
https://doi.org/10.1111/poms.12017.
Bruccoleri, M, Cannella, S & La Porta, G 2014, ‘Inventory record inaccuracy in supply
chains: the role of workers’ behavior’, International Journal of Physical Distribution and
Logistics Management, vol. 44, no. 10, pp. 796–819, https://doi.org/10.1108/IJPDLM-09-
2013-0240.
Bryman, A 2012, Social research methods, Oxford University Press, New York.
Burbidge, JL 1961, ‘The new approach to production’, Production Engineer, vol. 40, no. 12,
pp. 769–84, https://doi.org/10.1049/tpe.1961.0104.
Burbidge, JL 1984, ‘Automated production control with a simulation capacity’, Proceedings
of IFIP WG 5.7 Working Conference on Modelling Production Management Systems,
Copenhagen, pp. 19–35.
Byrne, KA, Silasi-Mansat, CD & Worthy, DA 2015, ‘Who choked under pressure? The big
five personality traits and decision-making under pressure’, Personality and Individual
Differences, vol. 74, pp. 22–8, https://doi.org/10.1016/j.paid.2014.10.009.
Cachon, GP & Lariviere, MA 1999, ‘Capacity allocation using past sales: when to turn-and-
earn’, Management Science, vol. 45, no. 5, pp. 685–703,
https://doi.org/10.1287/mnsc.45.5.685.
Cannella, S, Di Mauro, C, Dominguez, R, Ancarani, A & Schupp, F 2019, ‘An exploratory
study of risk aversion in supply chain dynamics via human experiment and agent-based
simulation’, International Journal of Production Research, vol. 57, no. 4, pp. 985–99,
https://doi.org/10.1080/00207543.2018.1497817.
Cantor, DE & Katok, E 2012, ‘Production smoothing in a serial supply chain: w laboratory
investigation’, Transportation Research Part E: Logistics and Transportation Review, vol.
48, no. 4, pp. 781–94, https://doi.org/10.1016/j.tre.2012.01.005.
Cantor, DE & Macdonald, JR 2009, ‘Decision-making in the supply chain: examining
problem solving approaches and information availability’, Journal of Operations
Management, vol. 27, no. 3, pp. 220–32, https://doi.org/10.1016/j.jom.2008.09.002.
Cao, Q, Baker, J & Schniederjans, D 2014, ‘Bullwhip effect reduction and improved business
performance through guanxi: an empirical study’, International Journal of Production
Economics, vol. 158, pp. 217–30, https://doi.org/10.1016/j.ijpe.2014.07.030.
Chartered Association of Business Schools 2018, Academic journal guide 2018, London.
Costantino, F, Gravio, GD, Shaban, A & Tronci, M 2014, ‘SPC-based inventory control
policy to improve supply chain dynamics’, International Journal of Engineering and
Technology, vol. 6, no. 1, pp. 418–26.
Croson, R & Donohue, K 2003, ‘Impact of POS data sharing on supply chain management:
An experimental study’, Production and Operations Management, vol. 12, no. 1, pp. 1–11,
https://doi.org/10.1111/j.1937-5956.2003.tb00194.x.
Croson, R & Donohue, K 2005, ‘Upstream versus downstream information and its impact on
the bullwhip effect’, System Dynamics Review, vol. 21, no. 3, pp. 249–60,
https://doi.org/10.1002/sdr.320.
Croson, R & Donohue, K 2006, ‘Behavioral causes of the bullwhip effect and the observed
value of inventory information’, Management Science, vol. 52, no. 3, pp. 323–36,
https://doi.org/10.1287/mnsc.1050.0436.

60
Croson, R, Donohue, K, Katok, E & Sterman, J 2014, ‘Order stability in supply chains:
coordination risk and the role of coordination stock’, Production and Operations
Management, vol. 23, no. 2, pp. 176–96, https://doi.org/10.1111/j.1937-5956.2012.01422.x.
Dean, JW Jr & Sharfman, MP 1993, ‘The relationship between procedural rationality and
political behavior in strategic decision-making’, Decision Sciences, vol. 24, no. 6, pp. 1069–
83, https://doi.org/10.1111/j.1540-5915.1993.tb00504.x.
Delhoum, S & Scholz-Reiter, B 2009, ‘The influence of decision patterns of inventory control
on the bullwhip effect based on a simulation game of a production network’, Production
Planning and Control, vol. 20, no. 8, pp. 666–77,
https://doi.org/10.1080/09537280902917757.
Dessí, R & Zhao, X 2018, ‘Overconfidence, stability and investments’, Journal of Economic
Behavior & Organization, vol. 145, pp. 474–94, https://doi.org/10.1016/j.jebo.2017.11.030.
Di Mauro, C, Ancarani, A, Schupp, F & Croco, G 2020, ‘Risk aversion in the supply chain:
evidence from replenishment decisions’, Journal of Purchasing and Supply Management, vol.
26, no. 4, 100646, https://doi.org/10.1016/j.pursup.2020.100646.
Diehl, E & Sterman, JD 1995, ‘Effects of feedback complexity on dynamic decision making’,
Organizational Behavior and Human Decision Processes, vol. 62, no. 2, pp. 198–215,
https://doi.org/10.1006/obhd.1995.1043.
Disney, SM, Towill, DR & Van De Velde, W 2004, ‘Variance amplification and the golden
ratio in production and inventory control’, International Journal of Production Economics,
vol. 90, no. 3, pp. 295–309, https://doi.org/10.1016/j.ijpe.2003.10.009.
Donohue, K & Croson, R 2002, ‘Experimental economics in practice’, Interfaces, vol. 32, no.
5, pp. 74–82, https://doi.org/10.1287/inte.32.5.74.37.
Durach, CF, Kembro, J & Wieland, A 2017, ‘A new paradigm for systematic literature
reviews in supply chain management’, Journal of Supply Chain Management, vol. 53, no. 4,
pp. 67–85, https://doi.org/10.1111/jscm.12145.
Edali, M & Yasarcan, H 2016, ‘Results of a beer game experiment: should a manager always
behave according to the book?’, Complexity, vol. 21, pp. 190–99,
https://doi.org/10.1002/cplx.21731.
Eksoz, C, Mansouri, SA & Bourlakis, M 2014, ‘Collaborative forecasting in the food supply
chain: a conceptual framework’, International Journal of Production Economics, vol. 158, pp.
120–35, https://doi.org/10.1016/j.ijpe.2014.07.031.
Fahimnia, B, Pournader, M, Siemsen, E, Dendoly, E, & Wang, C 2019, ‘Behavioral
operations and supply chain management: a review and literature mapping’, Decision
Sciences, vol. 50, no. 6, pp. 1127–83, https://doi.org/10.1111/deci.12369.

61
Fahimnia, B, Sander, N, & Siemsen, E 2020, ‘Human judgment in supply chain forecasting’,
Omega, 94, 102249, https://doi.org/10.1016/j.omega.2020.102249.
Feng, T, Keller, LR & Zheng, X 2011, ‘Decision making in the newsvendor problem: a cross-
national laboratory study’, Omega: The International Journal of Management Science, vol.
39, no.1, pp. 41–50, https://doi.org/10.1016/j.omega.2010.02.003.
Forrester, JW 1958, ‘Industrial dynamics: a major breakthrough for decision makers’,
Harvard Business Review, vol. 36, pp. 37–66.
Forrester, JW 1961, Industrial dynamics, MIT Press, Cambridge.
Fu, D, Ionescu, CM, Aghezzaf, E-H & De Keyser, R 2014, ‘Decentralized and centralized
model predictive control to reduce the bullwhip effect in supply chain management’,
Computers and Industrial Engineering. vol. 73, no. 1, pp. 21–31,
https://doi.org/10.1016/j.cie.2014.04.003.
Gavirneni S & Xia Y 2009, ‘Anchor selection and group dynamics in newsvendor decisions:
a note’, Decision Analysis, vol. 6, no. 2, pp. 87–97, https://doi.org/10.1287/deca.1090.0136.
Geary, S, Disney, SM & Towill, DR 2006, ‘On bullwhip in supply chains: historical review,
present practice and expected future impact’, International Journal of Production Economics,
vol. 101, no. 1, pp. 2–18, https://doi.org/10.1016/j.ijpe.2005.05.009.
Giard, V & Sali, M 2013, ‘The bullwhip effect in supply chains: a study of contingent and
incomplete literature’, International Journal of Production Research, vol. 51, no. 13, pp.
3880–93, https://doi.org/10.1080/00207543.2012.754552.
Gino, F & Pisano, G 2008, ‘Toward a theory of behavioral operations’, Manufacturing &
Service Operations Management, vol. 10, no. 4, pp. 676–91,
https://doi.org/10.1287/msom.1070.0205.
Haines, R, Hough, J & Haines, D 2017, ‘A metacognitive perspective on decision making in
supply chains: revisiting the behavioral causes of the bullwhip effect’, International Journal
of Production Economics, vol. 184, pp. 7–20, https://doi.org/10.1016/j.ijpe.2016.11.006.
Haines, R, Hough, JR & Haines, D 2010, ‘Individual and environmental impacts on supply
chain inventory management: an experimental investigation of information availability and
procedural rationality’, Journal of Business Logistics, vol. 31, no. 2, pp. 111–28,
https://doi.org/10.1002/j.2158-1592.2010.tb00144.x.
Hammond, J 1994, Barilla SpA (A), case study no. 694046, Harvard Business School
Publishing, Boston.
Hart, C 1998, Doing a literature review: releasing the social science research imagination,
Sage, Thousand Oaks.

62
Hofmann, E 2017, ‘Big data and supply chain decisions: the impact of volume, variety and
velocity properties on the bullwhip effect’, International Journal of Production Research, vol.
55, no. 17, pp. 5108–26, https://doi.org/10.1080/00207543.2015.1061222.
Hung, K-T & Ryu, S 2008, ‘Changing risk preferences in supply chain inventory decisions’,
Production Planning and Control, vol. 19, no. 8, pp.770–80,
https://doi.org/10.1080/09537280802550146.
Hung, W-H, Lin, C-P & Ho, C-F 2014, ‘Sharing information in a high uncertainty
environment: lessons from the divergent differentiation supply chain’, International Journal
of Logistics Research and Applications, vol. 17, no. 1, pp. 46–63,
https://doi.org/10.1080/13675567.2013.837156.
Jin, HW 2015, ‘Analysis of student behavior in the beer game’, Information (Japan), vol. 18,
no. 6, pp. 2275–82.
Johnson-Laird, PN 1983, Mental models: towards a cognitive science of language, inference,
and consciousness, Cambridge, MA: Harvard University Press.
Jones, O & Gatrell, C 2014, ‘Editorial: the future of writing and reviewing for IJMR’,
International Journal of Management Reviews, vol. 16, no. 3, pp. 249–64,
https://doi.org/10.1111/ijmr.12038.
Khan, MH, Ahmed, S & Hussain, D 2019, ‘Analysis of bullwhip effect: a behavioral
approach’, Supply Chain Forum, vol. 20, no. 4, pp. 310–31,
https://doi.org/10.1080/16258312.2019.1661756.
Kim, JG, Chatfield, D, Harrison, TP & Hayya, JC 2006, ‘Quantifying the bullwhip effect in a
supply chain with stochastic lead time’, European Journal of Operational Research, vol. 173,
no. 2, pp. 617–36, https://doi.org/10.1016/j.ejor.2005.01.043.
Kovacevic, I, Panic, B, Vujosevic, M & Kuzmanovic, M 2013, ‘Application of transactional
analysis in bullwhip effect analysis’, Amfiteatru Economic, vol. 15, no. 33, pp. 210–23.
Lee, HL, Padmanabhan, V & Whang, SJ 1997, ‘Information distortion in a supply chain: the
bullwhip effect’, Management Science, vol. 43, no. 4, pp. 546–58,
https://doi.org/10.1287/mnsc.43.4.546.
Li, Z & Yan, G 2015, ‘Exploring different order decision behaviors with bullwhip effect and
service level measures in supply chain system’, Discrete Dynamics in Nature and Society,
vol. 2015, no. 6, pp. 1–8, https://doi.org/10.1155/2015/657352.
Lin, J, Naim, MM, Purvis, L & Gosling, J 2017, ‘The extension and exploitation of the
inventory and order based production control system archetype from 1982 to 2015’,
International Journal of Production Economics, vol. 194, pp. 135–52,
https://doi.org/10.1016/j.ijpe.2016.12.003.

63
Loch, CH & Wu, Y 2005, ‘Behavioral operations management’, Foundations and Trends in
Technology, Information and Operations Management, vol. 1, no. 3, pp. 121–232,
https://doi.org/10.1561/0200000009.
Liu, H, Howley, E & Duggan, J 2012, ‘Co-evolutionary analysis: a policy exploration method
for system dynamics models’, System Dynamics Review, vol. 28, no. 4, pp. 361–9,
https://doi.org/10.1002/sdr.1482.
Macdonald, JR, Frommer, ID & Karaesmen, IZ 2013, ‘Decision making in the beer game and
supply chain performance’, Operations Management Research, vol. 6, no. 3–4, pp. 119–26,
https://doi.org/10.1007/s12063-013-0083-4.
Machuca, JAD & Barajas, RP 2004, ‘The impact of electronic data interchange on reducing
bullwhip effect and supply chain inventory costs’, Transportation Research Part E, vol. 40,
no. 3, pp. 209–28, https://doi.org/10.1016/j.tre.2003.08.001.
Mazur, JE 2006, ‘Mathematical models and the experimental analysis of behaviour’, Journal
of the Experimental Analysis of Behaviour, vol. 82, no. 2, pp. 275–91,
https://doi.org/10.1901/jeab.2006.65-05.
Miragliotta, G 2006, ‘Layers and mechanisms: a new taxonomy for the bullwhip effect’,
International Journal of Production Economics, vol. 104, pp. 365–81,
https://doi.org/10.1016/j.ijpe.2005.10.001.
Moon, S-A & Kim, D-J 2005, ‘Systems thinking ability for supply chain management’,
Supply Chain Management, vol. 10, no. 5, pp. 394–401,
https://doi.org/10.1108/13598540510624214.
Naim, MM, Spiegler, VL, Wikner, J & Towill, DR 2017, ‘Identifying the causes of the
bullwhip effect by exploiting control block diagram manipulation with analogical reasoning’,
European Journal of Operational Research, vol. 263, no. 1, pp. 240–6,
https://doi.org/10.1016/j.ejor.2017.05.014.
Narayanan, A & Moritz, BB 2015, ‘Decision making and cognition in multi-echelon supply
chains: an experimental study’, Production and Operations Management, vol. 24, no. 8, pp.
1216–34, https://doi.org/10.1111/poms.12343.
Ng, S, Lee, J & Soutar, G 2007, ‘Are Hofstede’s and Schwartz’s value frameworks
congruent?’, International Marketing Review, vol. 24, pp. 164–80,
https://doi.org/10.1108/02651330710741802.
Nienhaus, J, Ziegenbein, A & Schoensleben, P 2006, ‘How human behaviour amplifies the
bullwhip effect: a study based on the beer distribution game online’, Production Planning
and Control, vol. 17, no. 6, pp. 547–57, https://doi.org/10.1080/09537280600866587.

64
Niranjan, TT, Metri, BA & Aggarwal, V 2009, ‘The behavioural causes of the bullwhip
effect: breaking the mould’, International Journal of Services and Operations Management,
vol. 5, no. 3, pp. 350–74, https://doi.org/10.1504/IJSOM.2009.024151.
Niranjan, TT, Wagner, SM & Bode, C 2011, ‘An alternative theoretical explanation and
empirical insights into overordering behavior in supply chains’, Decision Sciences, vol. 42,
no. 4, pp. 859–88, https://doi.org/10.1111/j.1540-5915.2011.00334.x.
Pamulety, T & Pillai, M 2016, ‘Effect of customer demand information sharing on a four-
stage serial supply chain performance: an experimental study’, Uncertain Supply Chain
Management, vol. 4, pp. 1–16, https://doi.org/10.5267/j.uscm.2015.10.001.
Perera, HN, Hurley, J, Fahimnia, B & Reisi, M 2019, ‘The human factor in supply chain in
supply chain forecasting: a systematic review’, European Journal of Operational Research,
vol. 274, no. 2, pp. 574–600, https://doi.org/10.1016/j.ejor.2018.10.028.
Perera, HN, Fahimnia, B & Tokar, T 2020, ‘Inventory and ordering decisions: a systematic
review on research driven through behavioral experiments’, International Journal of
Operations and Production Management, vol. 40, no.7-8, pp. 997–1039,
https://doi.org/10.1108/IJOPM-05-2019-0339.
Podrug, N 2011, ‘Influence of national culture on decision-making style’, South East
European Journal of Economics and Business, vol. 6, pp. 37–44,
https://doi.org/10.2478/v10033-011-0004-0.
Poornikoo, M & Qureshi, MA 2019, ‘System dynamics modeling with fuzzy logic application
to mitigate the bullwhip effect in supply chains’, Journal of Modelling in Management, vol.
14, no. 3, pp. 610–27, https://doi.org/10.1108/JM2-04-2018-0045.
Riddalls, CE & Bennett, S 2003, ‘Quantifying bounded rationality: managerial behaviour and
the Smith predictor’, International Journal of Systems Science, vol. 34, no. 6, pp. 413–26,
https://doi.org/10.1080/00207720310001614097.
Rong, Y, Shen, Z-J-M & Snyder, LV 2008, ‘The impact of ordering behavior on order-
quantity variability: a study of forward and reverse bullwhip effects’, Flexible Services and
Manufacturing Journal, vol. 20, no. 1–2, pp. 95–124, https://doi.org/10.1007/s10696-009-
9054-3.
Sadeghi, J, Mousavi, SM, Niaki, STA & Sadeghi, S 2013, ‘Optimizing a multi-vendor multi-
retailer vendor managed inventory problem: two tuned meta-heuristic algorithms’,
Knowledge-Based Systems. vol. 50, pp.159–70, https://doi.org/10.1016/j.knosys.2013.06.006.
Sarkar, S & Kumar, S 2015, ‘A behavioral experiment on inventory management with supply
chain disruption’, International Journal of Production Economics, vol. 169, pp. 169–78,
https://doi.org/10.1016/j.ijpe.2015.07.032.

65
Saunders, M, Lewis, P & Thornhill, A 2009, Research methods for business students, 5th edn,
Pearson Education, Harlow.
Schacter, DL, Gilbert, DT & Wegner, DM 2011, Psychology, New York: Worth.
Schramm, NJ 2001, ‘Cultural dimensions of decision making: Denmark and France
compared’, Journal of Managerial Psychology, vol. 16, no. 6, pp. 404–23,
https://doi.org/10.1108/02683940110402389.
Shee, H & Kaswi, S 2016, ‘Behavioral causes of the bullwhip effect: multinational vs. local
supermarket retailers’, Operations and Supply Chain Management—An International Journal,
vol. 9, no. 1, pp. 1–14, https://doi.org/10.31387/oscm0230156.
Simon, H 1947, Administrative behavior: a study of decision-making processes in
administrative organization, The Macmillan Company, New York.
Simon, H 1957, Models of man, Wiley, New York, https://doi.org/10.2307/2550441.
Simon, H 1983, Reason in human affairs, Stanford University, Stanford.
Snyder, H 2019, ‘Literature review as a research methodology: an overview and guidelines’,
Journal of Business Research, vol. 104, pp. 333–9,
https://doi.org/10.1016/j.jbusres.2019.07.039.
Spina, G, Caniato, F, Luzzini, D & Ronchi, S 2013, ‘Past, present and future trends of
purchasing and supply management: an extensive literature review’, Industrial Marketing
Management, vol. 42, no. 8, pp. 1202–12, https://doi.org/10.1016/j.indmarman.2013.04.001.
Steckel, JH, Gupta, S & Banerji, A 2004, ‘Supply chain decision making: will shorter cycle
times and shared point-of-sale information necessarily help?’, Management Science, vol. 50,
no. 4, pp. 458–64, https://doi.org/10.1287/mnsc.1030.0169.
Sterman, JD 1988, ‘Deterministic chaos in models of human behavior: methodological issues
and experimental results’, System Dynamics Review, vol. 4, no. 1–2, pp. 148–78,
https://doi.org/10.1002/sdr.4260040109.
Sterman, JD 1989a, ‘Modeling managerial behaviour: misperceptions of feedback in a
dynamic decision-making experiment’, Management Science, vol. 35, no. 3, pp. 321–39,
https://doi.org/10.1287/mnsc.35.3.321.
Sterman, JD 1989b, ‘Misperceptions of feedback in dynamic decision making’,
Organizational Behavior and Human Decision Processes, vol. 43, no. 3, pp. 301–35,
https://doi.org/10.1016/0749-5978(89)90041-1.
Sterman, JD 1994, ‘Learning in and about complex systems’, System Dynamics Review, vol.
10, pp. 291–330, https://doi.org/10.1002/sdr.4260100214.

66
Sterman, JD 2000, Business dynamics, system thinking and modeling for a complex world,
McGraw Hill, New York.
Sterman, JD 2006, ‘Learning from evidence in a complex world’, American Journal of Public
Health, vol. 96, no. 3, pp. 505–14, https://doi.org/10.2105/AJPH.2005.066043.
Sterman, JD & Dogan, G 2015, ‘“I’m not hoarding, I’m just stocking up before the hoarders
get here”: behavioral causes of phantom ordering in supply chains’, Journal of Operations
Management, vol. 39–40, pp. 6–22, https://doi.org/10.1016/j.jom.2015.07.002.
Stuart, I, Mccutcheon, D, Handfield, R, McLachlin, R & Samson, D 2002, ‘Effective case
research in operations management: a process perspective’, Journal of Operations
Management, vol. 20, pp. 419–33, https://doi.org/10.1016/S0272-6963(02)00022-0.
Su, X 2008, ‘Bounded rationality in newsvendor models’, Manufacturing and Service
Operations Management, vol. 10, no. 4, pp. 566–89,
https://doi.org/10.1287/msom.1070.0200.
Syntetos, AA, Georgantzas, NC, Boylan, JE & Dangerfield, BC 2011, ‘Judgement and supply
chain dynamics’, Journal of the Operational Research Society, vol. 62, no. 6, pp. 1138–58,
https://doi.org/10.1057/jors.2010.56.
Thompson, ER 2008, ‘Development and validation of an international English Big-five Mini-
makers’, Personality and Individual Differences, vol. 45, no. 6, pp. 542–48,
https://doi.org/10.1016/j.paid.2008.06.013.
Thompson, M & Badizadegan, D 2015, ‘Valuing information in complex systems: an
integrated analytical approach to achieve optimal performance in the beer distribution game’,
IEEE Access, vol. 3, pp. 2677–86, https://doi.org/10.1109/ACCESS.2015.2505730.
Tokar, T, Aloysius, J & Waller, M 2012, ‘Supply chain inventory replenishment: the
debiasing effect of declarative knowledge’, Decision Sciences, vol. 43, no. 3, pp. 525–46,
https://doi.org/10.1111/j.1540-5915.2012.00355.x.
Tokar, T, Aloysius, J, Waller, M & Hawkins, DL 2016, ‘Exploring framing effects in
inventory control decisions: violations of procedure invariance’, Production and Operations
Management, vol. 25, no. 2, pp. 306–29, https://doi.org/10.1111/poms.12430.
Tranfield, D, Denyer, D & Smart, P 2003, ‘Towards a methodology for developing evidence-
informed management knowledge by means of systematic review’, British Journal of
Management, vol. 14, no. 3, pp. 207–22, https://doi.org/10.1111/1467-8551.00375.
Turner, BL, Goodman, M, Machen, R, Mathis, C, Rhoades R & Dunn B 2020, ‘Results of
beer game trials played by natural resource managers versus students: does age influence
ordering decisions?’, Systems, vol. 8, no. 4, pp. 1–30,
https://doi.org/10.3390/systems8040037.

67
Udenio, M, Fransoo, JC & Peels, R 2015, ‘Destocking, the bullwhip effect, and the credit
crisis: empirical modeling of supply chain dynamics’, International Journal of Production
Economics, vol. 160, pp. 34–46, https://doi.org/10.1016/j.ijpe.2014.09.008.
Udenio, M, Vatamidou, E, Fransoo, JC & Dellaert, N 2017, ‘Behavioral causes of the
bullwhip effect: an analysis using linear control theory’, IISE Transactions, vol. 49, no. 10,
pp. 980–1000, https://doi.org/10.1080/24725854.2017.1325026.
Van Engelenburg, S, Janssen, M & Klievink, B (2018), ‘A blockchain architecture for
reducing the bullwhip effect’, Business Modeling and Software Design, pp. 69–82,
https://doi.org/10.1007/978-3-319-94214-8_5.
Villa, S, Gonçalves, P & Arango, S 2015, ‘Exploring retailers’ ordering decisions under
delays’, System Dynamics Review, vol. 31, no. 1–2, pp. 1–27,
https://doi.org/10.1002/sdr.1527.
Von Lanzenauer, CH & Pilz-Glombik, K 2002, ‘Coordinating supply chain decisions: an
optimization model’, OR Spectrum, vol. 24, no. 1, pp. 59–78, https://doi.org/10.1007/s291-
002-8200-3.
Wang, X & Disney, SM 2016, ‘The bullwhip effect: progress, trends and directions’,
European Journal of Operational Research, vol. 250, no. 3, pp. 691–701,
https://doi.org/10.1016/j.ejor.2015.07.022.
Wiedenmann, M & Größler, A 2019, ‘The impact of digital technologies on operational cause
of bullwhip effect: a literature review’, Procedia CIRP, vol. 80, pp. 552–7,
https://doi.org/10.1016/j.procir.2019.03.154.
Wu, DY & Katok, E 2006, ‘Learning, communication, and the bullwhip effect’, Journal of
Operations Management, vol. 24, no. 6, pp. 839–50,
https://doi.org/10.1016/j.jom.2005.08.006.
Zhao, X, Huo, B, Flynn, BB & Yeung, JHY (2008), ‘The impact of power and relationship
commitment on the integration between manufacturers and customers in a supply chain’,
Journal of Operations Management, vol. 26, no. 3, pp. 368–88,
https://doi.org/10.1016/j.jom.2007.08.002.
Zhao, Y & Zhao, X 2015, ‘On human decision behavior in multi-echelon inventory
management’, International Journal of Production Economics, vol. 161, pp. 116–28,
https://doi.org/10.1016/j.ijpe.2014.12.005.

68