Proceedings of IDETC/CIE 2006

ASME 2006 International Design Engineering Technical Conferences

& Computers and Information in Engineering Conference
September 10-13, 2006, Philadelphia, Pennsylvania, USA

DETC2006-99229

PLATFORM-BASED DESIGN AND DEVELOPMENT:

CURRENT TRENDS AND NEEDS IN INDUSTRY
Timothy W. Simpson and Tucker Marion Olivier de Weck
The Pennsylvania State University Massachusetts Institute of Technology
University Park, PA Cambridge, MA
{tws8, tjm342}@psu.edu [email protected]

Katja Hölttä-Otto Michael Kokkolaras Steven B. Shooter

University of Massachusetts – Dartmouth University of Michigan Bucknell University
North Dartmouth, MA Ann Arbor, MI Lewisburg, PA
[email protected] [email protected] [email protected]

ABSTRACT of economies of scale and scope while satisfying a variety of

Many companies constantly struggle to find cost-effective market segments. Successful examples can be found in a variety
solutions to satisfy the diverse demands of their customers. In of companies, including Black & Decker [1], Seagate [2], Sony
this paper, we report on two recent industry-focused [3], and Volkswagen [4] to name a few. Planning families of
conferences that emphasized platform design, development, and products requires additional care and attention, since each
deployment as a means to increase variety, shorten lead-times, product competes for market share not only with competitor
and reduce development and production costs. The first products, but also with other products in the family.
conference, Platform Management for Continued Growth, was A product family is a group of related products that are
held November-December 2004 in Atlanta, Georgia, and the derived from a common set of components, modules, and/or
second, 2005 Innovations in Product Development Conference subsystems to satisfy a variety of market niches. The key to a
- Product Families and Platforms: From Strategic Innovation successful product family is the product platform around which
to Implementation, was held in November 2005 in Cambridge, the product family is derived [5]. Product platform definitions
Massachusetts. The two conferences featured presentations range from the “set of common components, modules, or parts
from academia and more than 20 companies who shared their from which a stream of derivative products can be efficiently
successes and frustrations with platform design and deployment, developed and launched” [5] to the “collection of assets [i.e.,
platform-based product development, and product family components, processes, knowledge, people and relationships]
planning. Our intent is to provide a summary of the common that are shared by a set of products” [6]. Some industries view
themes that we observed in these two conferences. Based on platforms at a more abstract level, defining not only the set of
this discussion, we extrapolate upon industry’s needs in common elements but also the architectural rules that enable a
platform design, development, and deployment to stimulate and set of planned product offerings where the architectural rules
catalyze future work in this important area of research. define geometrical, mechanical, electrical, and software
interfaces between platform elements [7,8]. Firms developing
Keywords: Product Platform, Product Family, Product Variety, infrastructures for oil [9] and space exploration [10], for
Modularity, Commonality example, will continue to expand and evolve this definition as
they seek to exploit the benefits of using platforms.
1. INTRODUCTION Designing a product platform and corresponding family of
Marketplace globalization, the proliferation of niche products is a difficult task that embodies all of the challenges of
markets, increased competitive pressures, and demand for product design while adding the complexity of coordinating the
customized products have rendered the practice of isolated design of multiple products in an effort to increase commonality
design and production of individual products nearly obsolete. across the set of products without compromising their
Across many industries, the prevailing practice is to design distinctiveness. Due to their difficulty, product family design
families of products that exploit commonality to take advantage and platform-based product development have been primarily

1 Copyright © 2006 by ASME

practiced in an ad hoc fashion. Academic research efforts Wireless, Argon Engineering, and Innovation Focus. Patrick
initiated about a decade ago sought to develop systematic Gordon from PRTM hosted a discussion-filled post-conference
methodologies for qualitative and quantitative product platform workshop entitled, Tapping the Full Potential of Product
and product family design. The former address pertinent Platforms, which helped coalesce the previous two day’s talks.
research issues from a business-oriented perspective while the Several common themes arose that were applicable to the new
latter focus on the engineering aspect. Product platform and product development processes for both goods and services as
family design has become a very active and increasingly discussed in Section 3. The workshop agenda and list of
relevant research topic, with its own share of special sessions in speakers can be found at: http://www.iirusa.com/platform/.
conferences and archival journals. A broad survey of existing
methodologies can be found in Ref. [11], and an in-depth 2.2. Product Families and Platforms: From Strategic
discussion of many of these methods can be found in Ref. [12]. Innovation to Implementation
As this research field has matured, it has attracted increased The second platform-oriented conference was held at the
attention from industry, which has spurred renewed interest in Massachusetts Institute of Technology (MIT) on November 9-
academia. This has led to many industry-funded projects and 10, 2005. In total, 114 individuals participated, split almost
case studies that demonstrated the potential of the developed evenly between industry and academia, including 20 students
methodologies. At the same time, application of academic from various universities in the U.S., Europe, and Asia. There
efforts on real-world problems exposed limitations and needs were three primary objectives for this conference:
for further research. We are encouraged by the two-way 1. bring together a community of practitioners and academics
feedback process that is starting to occur between industry and to learn, think, debate and discuss the latest trends and
academia. The two industry-focused conferences reviewed in achievements in platform-based product family design;
the next section are the most recent examples of balanced 2. present state-of-the-art methods and tools for product
interactions among academia and industry, where such platform and product family design, coincident with the
challenges were identified along with additional potential release of a new edited volume on the topic [12]; and
sectors that may benefit from adopting product family design 3. understand how the concept of platform-based product
strategies and platform-based development approaches. Our family design can be extended to new areas such as
intent is thus to report on these opportunities, identify common services and software, beyond the traditional focus on
themes, and elaborate on future research needs and challenges electro-mechanical products.
to stimulate further growth in this important area of research. The complete conference agenda and speaker presentations can
The remainder of this paper is as follows. In the next be found at: http://cipd.mit.edu/pd/.
section, we provide an overview of the two industry-focused The keynote speakers were Marc Meyer (Day 1) and B.
platform conferences that motivated this paper. In Section 3, we Joseph Pine II (Day 2). Meyer focused on the relationship
discuss the common themes and trends that arose from these between product platforms and the lifecycle phases of an
two conferences. Based on our observations, in Section 4 we enterprise (see Figure 1): early innovation, characterized by low
discuss the challenges and future research directions in sales volumes and technological discontinuities, the middle
platform-based design. Section 5 provides closing remarks. phase where some companies emerge as niche players, while
others experience rapid growth with concomitant market
2. OVERVIEW OF PLATFORM CONFERENCES expansion, followed by a third phase of either stagnation and
gradual decline or continued enterprise growth and
2.1. Platform Management for Continued Growth rejuvenation. He argued that managing the transitions between
Twenty industry experts convened in Atlanta, Georgia on the phases (gray vertical bars in Figure 1) was critical, and that
November 30 and December 1, 2004 for the first Platform product platforms, and more generally modular product
Management for Continued Growth conference to share architectures with well defined interfaces, play a critical role
strategies and results of their internal product platform design during those transitions. Rather than focusing on cost savings
and development efforts. The conference was co-organized by that can be achieved through commonality, future research
the Institute for International Research (IIR) and the Product should increasingly focus on new market applications, product
Development and Management Association (PDMA), and drew development speed and enterprise growth enabled by product
a small, but wide-ranging audience from industry and academia. platforms.
The conference began with a keynote speech by Marc On Day 2, Pine, the author of the Mass Customization: The
Meyer, co-author of The Power of Product Platforms: Building New Frontier in Business Competition [13], gave his vision of
Value and Cost Leadership [5]. He highlighted successes and future trends in the area of mass customization and how
failures, which he used as key aspects of platform development. platform-based development can support customized product
Other presentations were given by representatives from IBM, and service offerings. He also emphasized the important role of
Eastman Kodak, DuPont, Intel, and Lockheed Martin as well as modularity and product architecture in allowing companies to
mid- to small-sized companies such as Harley Davidson, best combine their capabilities and technologies to serve a wide
Playtex, Aventis Pasteur, Case-New Holland Global, Cingular variety of (changing) customer needs.

2 Copyright © 2006 by ASME

 and response to market shifts, reaching a zenith in 1993, when
corporate losses topped $8 Billion [15]. Symptomatic of the
organizational problems was their business server model line.
As Mugge [15] discussed, in 1996 several divisions were model
line-oriented, each producing uniquely ‘branded’ products. The
servers had upwards of 3,500 components; with divisions
having less than two percent of common, shared parts (any
reuse was unintentional). In response, IBM reorganized itself
into cross-disciplined market-facing platform teams, which
included marketing, sales, engineering, manufacturing, and
logistics. These teams have been designed to integrate and
master four core competencies:
1. increase ideation through Market Planning and Analysis,
Figure 1. Innovation and Enterprise Growth [14] 2. improve investment decisions from disciplined portfolio
management,
The remainder of the conference provided an overview of 3. increase innovation yield and flexibility from Platform
platform research and practice (Session I) and focused on Management, and
research trends and industry accomplishments in traditional 4. faster time-to-market through better Pipeline Management.
industries such as the automotive industry, industrial equipment, Since the reorganization, the eServer line has been developed
and appliances (Session II). In many of these industries, product and launched to critical and sustained sales success. Using
variety is increasingly achieved by modularizing products, common and preferred parts, there has been a 70-80% reduction
defining standard interfaces, and explicitly accommodating in part numbers, and over $700 Million eliminated from IBM’s
variety at later points in the assembly process (i.e., cost structure since the late 1990’s. Historical data, noting
postponement). Session III focused on recent expansions of the reduction in the number of components and cost reduction, are
platform concept into “non-traditional” areas such as the service shown in Figure 2. Additionally, Mugge stated that the number
industry and software product line architecture. Research is still of new products increased by 270% percent. By implementing
emerging in these areas, and specific examples of firms that cross-functional teams combined with platform management,
have successfully developed a modular reuse strategy (e.g., IBM has demonstrated the tangible benefits of culture change.
Aramark, HP Business Services) were given. The conference
ended with a panel session (IV) on the effects of globalization
on product development in general and product platforms
specifically. As the panelists commented, platforms enable firms
to offer global portfolios of products, while accounting for
regional differences in design, styling and regulations.

3. COMMON THEMES AND TRENDS

Several common themes and trends arose among the

presentations and examples discussed at the two conferences.
These themes present common challenges and solutions that can
be studied and utilized for the development of platforms and
platform-based goods and services. The following themes Figure 2. IBM P/N Reduction, P/N Reuse, and Cost
demonstrate how to succeed in leveraging the benefits of a Reduction [15]
holistic platform and platform-based development process.
3.2. Upper Management – Catalyst for Change
3.1. Corporate Culture Change Reorganization will fail without strong support from upper
Corporations that have embraced cross-functional product management. Corporations are difficult to ‘turn-on-a-dime,’ but
platform teams have routinely demonstrated the ability to changing heading is only possible if they have the means and
quickly reinvent themselves and successfully enter new markets. will to complete the necessary course corrections. IBM’s
Traditionally, corporations have employed functional reorganization, for example, produced dramatic results, but it
management structures. These individual ‘fiefdoms’ often have was only because IBM’s CEO at the time, Louis V. Gerstner,
their own R&D Centers, Manufacturing, and Supply Chain spearheaded the culture change by appointing senior
Management organizations. Little information, technology, or management to lead the effort and commit the required
business lessons are shared among the different business units. resources [16]. A related example of support for culture change
At IBM, this type of organization led to a lack of innovation occurred at Intel. In 2000, Intel’s Desktop Platform Group
strategy was changed from developing components to

3 Copyright © 2006 by ASME

developing platforms. Since that time, management has fully fundamentally new technologies, product platforms, and
supported evolutionary improvement and implementation of a customized variants are often very different. Some firms are
coherent Platform Development Management System. [17]. experimenting with layered organizational models where
The examples from IBM and Intel have a common attribute platform teams are acting as a connecting layer between the
of upper management integration into adopting platform slower science-related organizational layer of basic R&D and
management. Additionally, although IBM and Intel are entities the fast-paced market-related product development layer, where
with different corporate customs, each tailored a management designers are primarily concerned with tailoring and assembling
support and integration methodology that worked with their products from already existing and proven technologies and
existing culture. IBM appointed a change manager, while Intel components to respond quickly to changing customer demands.
fostered an evolutionary change environment. As such, each
corporation is unique, and how the change is realized will also 3.4. Architecture – Common Subsystems and Reduced
be unique. The commonality, however, begins with desire and Complexity
support of upper management, which is key if firms are to avoid Developing cohesive and flexible product architectures is a
stagnation and decline of their product portfolio. necessity in successfully implementing a platform strategy. The
platform should form the basis of an internal product roadmap
3.3. Product Development – Results through Teamwork that outlines future capabilities and functionality while serving
Cross-functional product development teams are essential as a pillar in the overall corporate vision [8]. Overall product
for implementing a successful platform development strategy. strategy is derived from the platform, as the platform should be
An example of a fully integrated approach is Sanofi Aventis, able to be tailored to meet different market segments and
developer and manufacturer of vaccines. Their product platform performance targets (see Figure 4). Platforms use supporting
team consists of representatives from R&D, manufacturing, elements such as common subsystems and components. This
marketing, quality assurance, logistics, and even the legal allows platforms to be designed for a particular market segment
department. The framework shown in Figure 3 allows Aventis to and then be easily modified for different segments and/or
develop early partnerships among team members and establish a higher-level tiers within the same segment.
smooth transition from research into the critical path of FDA
approval [18]. As McGill discussed, by tearing down the
developmental ‘walls’ and increasing communication between
the subsystems of the corporation, product development cycles
are reduced with the end result often being higher quality
products. Taking a step further, these cross-functional product
development teams can be aligned to have inter-team access to
common R&D, subsystems, and components. Cross-functional
teams have been used to great effect in other industries,
including automotive and aerospace.

Figure 4. Market Segmentation Grid [5]

Lockheed Martin has implemented a successful platform

strategy for its family of military transports [19]. With the next
series of Block Upgrades (major aircraft and avionics
revisions), software and avionics (major subsystems) will
become common for Lockheed’s three platforms, the C-130, C-
Figure 3. Shared Resources during Development [18] 5, and C-27. The common subsystems will form the basis for a
new airframe platform eventually replacing the C-130, which
IBM, Playtex, Intel, and DuPont all have aligned their will be similar to today’s aircraft in that is can be easily
organizations to maximize the benefits of cross-functional reconfigured to fill a wide variety of different roles. Today’s C-
teams and cross-organizational information sharing. Benefits 130 fills roles as diverse as Search and Rescue to Gunship
include component reduction, common architectures, and a applications, all leveraging a common airframe. Different
deep research pool to generate ‘out-of-the-box’ technology and software and weapons suites can be applied to the airframe to
ideas for innovative new products. Challenges, however, remain move up the Y-axis from ‘Low-cost’ to ‘High-end’ [19].
because the time constants involved in developing

4 Copyright © 2006 by ASME

 Modular platforming techniques are also finding new recognition-based platforms [21]: the first type allows
applications in telecommunications industries. Cingular companies to effectively utilize, and re-utilize a set of
Wireless is implementing a platform approach on Pre-paid and distinguishing features, characteristics, attributes or elements.
their ‘Take Charge’ cellular plans. These can leverage common The second allows them to efficiently utilize and re-utilize a set
technology subsystems such as wireless technology and be of product and/or process technologies, and the third one allows
applied to different demographic segments. Additionally, for (hopefully positive) brand recognition.
service plans and customer service can be easily reconfigured to
meet different market segment needs [20]. 3.6. Forecasting and Analysis – Understand the Market
In terms of planning a product platform-based product In order to develop successful products and services,
portfolio, there is consensus that an abundance of product corporations must accurately listen to and identify the needs and
variants is undesirable, both from the product lifecycle expectations of each market segment and tier. In looking at this
management’s (PLM’s) and from end-user’s/consumer’s point competitive landscape, each market niche needs to consider [8]:
of view. Therefore, complexity reduction constitutes a critical - What is the significance of this segment?
objective of the product family design process. Particularly, it is - What are the key products?
suggested that the number of product platforms is held to a - What are their volumes, revenue, and profits?
minimum to maximize commonality benefits, and that they are - What is the outlook for the next 5 years?
modular so that new product variants can be derived with - What does the Company have to do to enter, sustain, and
minimum effort and without having planned substantially for grow in the segment?
them. Moreover, it is recommended that existing product The company then develops a ‘360 degree’ view of potential
platforms are utilized to the maximum extent possible, since customers to understand their needs, requirements, and usage
their development requires a significant amount of resources. patterns. This ‘Voice of the Customer’ (VOC) approach has
This raises an interesting tension, since product platforms can been effective in helping guide the product specifications and
clearly increase product variety and short-term innovation; features of new product platforms.
however, because of the “sunk” investment into product A successful application of the VOC has occurred at Case-
platforms there is pressure to reuse them repeatedly. As pointed New Holland (CNH), a world leader in agricultural equipment
out by Pine, there may come a point where a platform acts as a such as tractors. In developing a new cross segment platform,
barrier to future innovation at which point renewal of the CNH embarked on an extensive program of interviewing
platform and underlying product architecture may be required. potential customers in each market. In person, one-on-one
interviews were held to gauge customer feedback on issues
3.5. Platform Strategies in “Non-Traditional” Applications ranging from cabin ergonomics to steering mechanisms.
As noted with Cingular Wireless, platforming techniques Responses were documented, analyzed, and used in the
are beginning to be used in other industries besides traditional conceptual development process to formulate product solutions.
product engineering firms. One of the consistent themes from This VOC process is integral to CNH’s process of Customer-
both conferences is that after a decade of research and Driven Product Definition [22]. Playtex also gave an example
development in the consumer products (e.g., electronics and of reorganizing a company in response to market needs [23].
home tooling) and engineering (e.g., automotive and aerospace)
sectors, platform-based product family design is now being
adopted in “non-traditional” sectors such as software
engineering, telecommunications, food and drugs industries,
and service systems (e.g., entertainment, tourism, banking).
Initially, this expansion occurs by marketing derivatives of
existing products and services to fill current and readily
exploited niches, but future product development will be
conducted using product platform strategies. Of utmost
importance for continued research and dialogue is having
different industries use common terminology. As observed by Figure 5. Requirements Allocation in Modern Systems
many representatives from diverse sectors, establishing Engineering [24]
common terminology may seem simple but is integral ingredient
to successful transfer of platform concepts to these non- Finally, it is becoming more widely accepted that product
traditional industries. The challenge of common terminology is family design approaches must be analytical and quantitative,
discussed further in Section 4. i.e., model-based (see Figure 5). One approach is to design
An interesting suggestion is that all of the following product platforms for robustness, i.e., insensitive to variations.
platform perspectives should be exploited to increase likelihood It has been suggested that this can be best accomplished by
of successful product launch: industrial and artistic design- using hierarchical and modular product architectures with
based platforms, technology-focused platforms, and brand “clean” interfaces to enable sensitivity analysis, error tracking,

5 Copyright © 2006 by ASME

statistical analysis of uncertainties and their propagation, and demand, revenue, and manufacturing costs are generally well
cascading of requirements and specifications that enable both understood, their detailed quantitative modeling remains
subcontractor flexibility and accountability [24,25]. It is elusive. The main reason is that as models of product
interesting to observe that hierarchical frameworks were performance, market demand, and manufacturing costs are
suggested for both traditional (engineering) and “non- concatenated, so are the modeling errors and uncertainties
traditional” (e.g., software engineering) applications [26]. inherent in them. To make matters worse, these errors are not
typically additive but multiplicative. It is thus imperative to
3.7. Financial Planning develop credible interdisciplinary product family development
Appropriate planning and architecture configuration for frameworks. Along with these, formal methods of model
product family development requires estimating expected validation and verification against engineering, market and cost
financial benefits both in terms of savings due to commonality data are needed. As many speakers at the two conferences
(manufacturing, inventory, training, maintenance) and revenues indicated, the engineering part of the framework already exists
due to successful product performance in the market. With only and/or may not be the hardest one to achieve; the rest, such as
a few exceptions, most existing methodologies for product integrated financial planning, is a challenge.
family design and development lack a rigorous cost-benefit
analysis: cost models and data are either not available or 3.8. Globalization and Product Platforms
proprietary, while expected product performance and revenues Finally, globalization not only offers opportunities for
are estimated using elementary net present value methods. Most product families but actually implies them. Consumers
methodologies are based on the implicit assumption that worldwide are increasingly wearing the same type of clothes,
maximized commonality is equivalent to maximized cost driving the same type of cars, working on the same type of
benefits. Even when cost models are included, they are used to computers using the same type of software. Additionally, people
quantify cost savings and to translate commonality to monetary are using the same type of appliances and telecommunication
units. Therefore, product commonality and differentiation is means, enjoying the same type of home entertainment, and
decided upon functional performance penalty (relative to playing the same type of electronic games. As highlighted by
products that do not share common parts or manufacturing many industry speakers, many of whom represent global
processes) considerations without taking into account losses or companies, product families are the only way for international
profits due to market performance. Attempts to quantify the companies to market their products efficiently and stay in
market impact of commonality in terms of demand and revenue business, as product families drive the competition. Figure 7
effects and to “close-the-loop” with the manufacturing savings depicts an example of how DuPont took a global perspective
(both fixed and variable costs) achieved through commonality while developing their platform strategy. In today’s global
are being made in both industry and academia. Figure 6 shows market, different regulations and cultural differences (not only
an end-to-end product modeling framework that maps key customer, but also within the firms and their various regional
platform commonality decisions through both the product divisions) must be taken into account. As such, product lines
architecture – engineering performance – product value-market- must be even more robust and flexible to such variations.
revenue path (upper) as well as the product architecture –
manufacturing cost – investment finance path (lower).
Technology Customers Competition
Performance Product
Attributes Value
-2- -3- -4-
Engineering f Product V Market
Performance Value Demand

x Soft
Attributes D P
Design
Variables s Demand Price

-1- -5- -6- A

Product Manufacturing C Investment Profit
Architecture Cost Finance NPV
Modules Variable
Parts Cost
Figure 7. Thinking Globally Offers New Opportunities to
Shareholders Exploit Platforms [28]
Labor Rates
Standards Raw Materials Interest Rates
Regulations Plants & Tooling
Suppliers 4. CHALLENGES AND FUTURE DIRECTIONS
Figure 6. Interdisciplinary Product Modeling Framework [27] A review of the common themes and trends leads directly
to identification of challenges industries face in deploying
While the general causal relationships between key platform strategies. Identification of these challenges creates
quantities such as commonality, product performance, market opportunities for future research as discussed next.

6 Copyright © 2006 by ASME

4.1. Formulation of Effective Industrial Partnerships strategies must recognize all of the elements described by
The companies that participated in the workshop were Robertson and Ulrich rather than focusing on just the physical
mostly the believers; that is, they had already embraced the aspects of components and processes. So, effective sharing of
concept of product platforms. However, they experienced knowledge and relationships are integral elements for realizing
challenges implementing the approach greater than anticipated an effective holistic platform strategy.
from the relatively straight-forward concept of “shared assets While technologies such as Enterprise Resource Planning
for leveraged benefit”. It is also clear that there is a disconnect (ERP) and Product Lifecycle Management (PLM) have made
between academia and industry. From the academic side, the inroads to supporting product development, there are currently
low level of application can be seen, for example, in how papers no tools available to facilitate the sharing of knowledge directed
on platforms tend to revert to the same, already dated, examples to product platforms [31,32]. Opportunities abound for
of platforms in industry. Similarly, many of the techniques and enhanced techniques for effectively capturing, storing,
tools from academia are not being applied in industry because retrieving and delivering information in support of product
they often do not scale well to complex or “messy” situations. platform strategies. Govindarajan from Hewlett-Packard
This disconnect could be remedied in several ways. The first is acknowledged a need to explore how documents can become
simply for academia to work more closely with industry on the primary vehicles for manipulating an information model in
research, and obviously more cooperative workshops bringing support of platforms [29]. This is just one facet of the broader
together both parties would help this. Another way to increase opportunities for knowledge management to support platforms.
the relevance and impact of academic platform research is to A common theme among industry representatives reflected
have students spend more time at host companies and for host the need for approaches in transforming the organization to
companies to make greater efforts in the area of data availability support platform strategies thereby highlighting the
and release. Other approaches could involve research consortia relationships aspect. Nidamarthi from ABB highlighted the
where non-competing firms from various industries who would importance of an organization aligned to implement and sustain
freely share data to jointly develop generalizable methods and the platform [33,34]. If the purpose of a company is to produce
tools to support platform-based design, development, and products to generate profit (based on platforms), then perhaps
decision-making. Another approach is to view this disconnect as the organization should be designed around the platform rather
a research opportunity. Is there a reason why certain platform than the other way around. This organizational change can be
design techniques and tools have been adopted in some difficult, possibly the biggest challenge as argued by Meyer
industries while others have not? What can we learn from this during the conference. Organization culture is not easy to
for developing future techniques and tools? change. Especially in the engineering literature, the organization
Another cause for disconnect identified was language, even that will develop the platform is often ignored. The challenge is
with the meaning of the term “platform”. The many different in how to get support and involvement from the entire
definitions of platforms create challenges for platform design. organization to this major change. There are industry examples
Different definitions among organizations can lead to tensions of success, such as IBM [15]; however, there are clearly ample
in common goals. For example, management may use the word opportunities for research into organizations, operations, and
platform for product lines and marketing may refer to customer human factors to support product platform strategies. Can
options as product modules, while engineering might call the diffusion of technology theories facilitate user adoption of
core technology of the company their platform. It is important platform techniques and tools?
for these multiple views to contribute to the same goal of
platform benefits in sync with the company strategy. The 4.3. Flexible Platform Design for Multiple Generations
challenge becomes how to define a platform, or how to combine As the lifetime of a platform is long with multiple product
the different definitions in a way that the entire organization can generations, one of the key challenges is to be able to predict
be on the same page. Govindarajan from Hewlett-Packard the future or to design the platforms so that the expected and
emphasized the need to explore how to generalize some of the unexpected changes can be accounted for during the original
core ideas of platforms along key dimensions such as portfolio, design of the platform. This calls for methods for designing
stakeholder, geography, and lifecycle so that they are more flexible product platforms. There is already work in the area
applicable to “non-traditional” areas such as service systems [35-37], but much more is needed. Key questions include:
[29]. There is an opportunity to establish a richer semantic Where to design platform flexibility? When to design a flexible
description for platforms to help identify these nuances. platforms and when to choose a platform update? How to
prepare for new applications, new product lines, and new
4.2. Recognizing a Holistic Platform Strategy radical technologies? How to properly value any flexibility
Considerable research has been conducted on techniques investments into platforms? When to initiate a complete
and strategies on formulating platforms based on physical platform redesign before it reaches stagnation and decline?
features, components or modules of products [11]. There has A related problem is determining the “extent” of the
also been considerable research on utilizing platforms for platform; see Ref. [27] and references therein. This refers to the
production processes [30]. It may be that effective platform question of how diverse the set of variants can be that is derived

7 Copyright © 2006 by ASME

from a common platform. The diversity can be quantified in reasons of policy or lack of enforcement of common standards.
terms of physical attributes such as geometry or in terms of We feel that the area of platform discipline represents a rich
functional performance. As a product platform is leveraged area for further research at the intersection of organizational
more-and-more over time, the capabilities (bandwidth) of the behavior and engineering design.
platform is constantly challenged with each new variant derived A common challenge in platform design is the management
from it [36]. One may choose to keep “extending” the platform of multiple tradeoffs. The possible benefits from a platform
or one may choose to strictly enforce its current configuration depend on the starting point for the platform strategy (e.g.,
and impose boundary conditions on new variants, in which case maximize commonality with minimal performance loss [40],
the variants may be overly constrained. If on the other hand one minimize cost, maximize variety). Surprisingly many methods
continues to continuously “stretch” the platform, it may claim a multitude of benefits but most only handle the tradeoff
eventually become overburdened, and it may be more efficient between two goals: performance and commonality [41]. This
to split the platform into two (or more) platforms. Thus, in some can result in abandoning the platform strategy, even though the
cases deploying multiple platforms may actually be optimal real problem lies in the misalignment of the goals and the
[37]. If the variety of functional requirements becomes too methods used to attain them. Ideally, a method would consider
large, the platform may become too demanding to develop, too cost, performance, variety, flexibility, etc. all at once, but this is
expensive to build and too complex to operate reliably. In that often unrealistic and intractable. Instead, typically a method
case it might be better to “descope” the platform and revert handles a tradeoff or two, leaving the remainder as separate
back to a collection of less ambitious, and simpler “special decisions. Trying to choose the best platform for a company is
purpose systems”. This is not purely an engineering or financial difficult. Learning about why different organizations strive for
question, but one of systems architecture and strategy. commonality reveals that the motivations can be quite varied. In
the oil industry, for example, the main reasons for increased
4.4. Corporate Platform Strategy and Tradeoffs standardization in oil platforms are not primarily capital
Platforms are related to the product architecture, supply expense savings from commonality but faster speed to “first oil”
chain, manufacturing, design reuse, etc. The platform strategy and higher levels of production reliability with fewer
should be considered not only as a part of a product strategy but interruptions due to diverse hardware [42].
also as a corporate strategy. Platform design can be the tool to As in single product development, in platform development
use to achieve the goals aligned in the company strategy. The too, the profits from the design take years to realize. The true
challenge is how to consider the full strategy in the success of a design can be objectively judged only at the end of
development i.e. how to take into account the multiple demands the product’s lifecycle. In platform development this problem is
of the entire strategy while designing the platform. In order to even more pronounced as a platform is designed to last for
implement a broad and effective platform strategy, substantial several product generations. The question becomes how to
management involvement is needed. But since the engineers are evaluate the “goodness” of a platform sooner, rather during the
typically the ones designing the platform, it is important also to development process already.
involve them to ensure that the crated platforms strategy During the MIT conference, PRTM showed an example,
guidelines will also be followed. Effective platform design where a medical device company applied a platform strategy
requires a truly company-wide effort. and saw positive results by reducing the total number of
Once the platform strategy is created and the platform(s) platforms being developed to lower the overall cost of
designed, the company faces a new problem – how to stick with development and speed up development [7]. In the automotive
the strategy and how to manage the platforms? A single industry, the success of a platform strategy is often described as
platform should carry over through multiple product a decrease in the number of different platforms and decrease in
generations, but how many, and how often should a single time-to-market. Oh [43] described how LG has benefited from
platform, or the entire platform strategy be updated? Should a their platform approach through cost reductions in
platform be adapted to changes when needed, or does that make manufacturing and development. Similarly, Marion discussed
the platform just a regular component that is redesigned as how Innovation Factory reduced the cost of development and
needed? Some researchers have addressed generational issues manufacturing as well as increased variety using a platform
such as these [3,38,39], but considerably more work is needed. approach [44]. It seems that the “goodness” of a platform can
Also, the phenomenon of platform discipline is related to this. be measured, but are these few measures enough to capture the
There is anecdotal evidence that the Joint Strike Fighter (JSF) financial success of the platform?
program started out with a relatively clean platform and a high Two companies, United Technologies [45] and PRTM [7],
degree of commonality between the carrier (U.S. Navy), addressed this issue in the MIT conference: both called for a
vertical-take-off (USMC), and USAF (long range) versions of multi-criteria approach to platform screening during platform
the aircraft, but that the amount of commonality between the development. Otto and Hölttä-Otto [46] suggest that companies
variants has gradually eroded. This erosion might happen for can use multiple metrics such as commonality, complexity,
legitimate engineering reasons – because the variants are more flexibility and reliability to evaluate which platform, among a
different than initially thought based on prototype tests – or for set of alternatives, a company should pursue. More work is

8 Copyright © 2006 by ASME

needed to define suitable metrics for specific cases and The most common themes resulting from the two
validation of how these metrics best reflect the future platform conferences are the need to think strategically about developing
success or failure. More research is needed to identify proper families of products or services and platforms based on
metrics for measuring platform success during the platform scalable, modular architectures with “clean”, standardized
evaluation phase as well as modeling platform and variant interfaces. In reality, however, “complete” modularity is not
performance. Rigorous models could be used to improve always fully achievable due to packaging, weight, power and
platform design and aid in selecting platform alternatives. volume constraints, among others. Quantifying both the benefits
and costs of platforming and standardization is necessary, but
4.5. Expanding Views for Platforms difficult due to inherent model and market uncertainties. It is
Platform concepts are expanding into new domains. Peter these uncertainties that also require platforms to be designed
McGrory [21] from the University of Art and Design Helsinki with robustness or flexibility to respond to future needs better.
discussed the importance of the relationship of the customer to These future needs could include new functional requirements
the product from an industrial design perspective. He stressed demanded by customers, new technologies, adherence to new
the importance of considering factors beyond the technology as regulations or the expansion into new geographical and
part of the platform strategy. While technology-related demographic markets. Product platforms tend to have lifetimes
intellectual property involves only patents, it is important to that exceed the lifetime of the variants that are derived from
consider other intellectual property such as trademarks, them and this makes the problem both challenging and relevant.
copyrights, and registered designs. An industrial design Finally, we are seeing expansion of platform concepts into
platform can be used to create brand image for products while new areas such as the service industry, software, large-scale
still benefiting from the common elements. Examining infrastructures, and military and other government systems. The
opportunities in this domain requires interdisciplinary starting point for success of platform concepts in these new
collaboration among industrial art, design and engineering and areas is an understandable and – generally – agreed upon
carry with it cross cultural and language challenges. nomenclature. We are encouraged by the interactions between
It is also clear that there is a need to recognize products industry and academia during past conferences and are
beyond physical artifacts. There is a need to explore the invigorated and confident that much research and
nuances of platform strategies applied to such areas as software implementation work remains to be undertaken in this
and services. Kathryn Weiss described her efforts to employ the interdisciplinary and fascinating research field.
product family approach to software development for spacecraft
control [26]. She stressed that success of software families is ACKNOWLEDGMENTS
contingent on developing and selecting an appropriate software This work was supported in part by the National Science
architecture to support the various products in the line; a similar Foundation under Grant Nos. DMI-0133923, IIS-0325402, and
approach was advocated by Harris [47]. There are clearly IIS-0325321; the Office of Naval Research under Grant No.
opportunities for exploring appropriate techniques from F015133; and by generous donations from the MIT Center for
physical product family planning in software architecture. Innovation in Product Development and DuPont. Any opinions,
Govindarajan [29] described HP’s approach to enabling mass findings, or recommendations in this work are the responsibility
customized services based on platform strategies. He identified of the authors and do not necessarily reflect the view of the
many of the same challenges in the service domain as in National Science Foundation or our collaborators and sponsors.
physical products; however, there clearly are opportunities for
research into the service sector that can be well received by our REFERENCES
advancing service economy. [1] Lehnerd, A. P., 1987, "Revitalizing the Manufacture and Design of Mature
Global Products," Technology and Global Industry: Companies and Nations in
the World Economy, B. R. Guile and H. Brooks, eds., National Academy Press,
5. CLOSING REMARKS
Washington, D.C., pp. 49-64.
Market pressures are forcing companies to rethink their [2] Brown, E., 2006, "Drive Fast, Drive Hard", Forbes, January 9, pp. 92
product development organization, develop new technologies, [3] Sanderson, S. W. and Uzumeri, M., 1997, Managing Product Families,
infuse these into platforms and derive customized variants from Irwin, Chicago, IL.
them. This encompasses the entire development process, from [4] Wilhelm, B., 1997, "Platform and Modular Concepts at Volkswagen - Their
Effect on the Assembly Process," Transforming Automobile Assembly:
market and customer research to supply chain management. Experience in Automation and Work Organization, K. Shimokawa, U. Jürgens
Integral to this change in a wide variety of industries is the and T. Fujimoto, eds., Springer, New York, pp. 146-156.
adoption of a platform management architecture. Successful [5] Meyer, M. H. and Lehnerd, A. P., 1997, The Power of Product Platforms:
traits among industry leaders are the formation of cross Building Value and Cost Leadership, Free Press, New York, NY.
[6] Robertson, D. and Ulrich, K., 1998, "Planning Product Platforms," Sloan
functional development teams, strong management support, Management Review, 39(4), pp. 19-31.
common platform architectures that maximize the sharing of [7] Bowman, D., 2005, "Platforming Trends in Industry," 2005 Innovations in
subsystems and components, and the ability to apply lessons Product Development Conference - Product Families and Platforms: From
learned over time for continuous improvement. Strategic Innovation to Implementation, Cambridge, MA.

9 Copyright © 2006 by ASME

[8] Gordon, P., 2004, "Tapping the Full Potential of Product Platforms: Best [31] Byron, B. M. and Shooter, S. B., 2005, "A Review of Software Solutions
Practices in Planning, Managing, and Organizing for Platform Effectiveness," for the Management of New Product Development and Product Family
Platform Management for Continued Growth, Atlanta, GA, IIR/PDMA. Planning," ASME Design Engineering Technical Conferences, Long Beach,
[9] Kalligeros, K., 2006, "Platforms and Real-Options in Large-Scale Systems," CA, ASME, Paper No. DETC2005/DAC-84454.
Ph.D. Dissertation, Engineering Systems Division, MIT, Cambridge, MA. [32] Shooter, S. B., Simpson, T. W., Kumara, S. R. T., Stone, R. B. and
[10] National Aeronautics and Space Administration, 2006, "Exploration Terpenny, J. P., 2005, "Toward a Multi-Agent Information Infrastructure for
Systems Architecture Study," http://www.nasa.gov/mission_pages/exploration/ Product Family Planning and Mass Customization," International Journal of
news/ESAS_report.html. Mass Customization, 1(1), pp. 134-155.
[11] Simpson, T. W., 2004, "Product Platform Design and Customization: [33] Nidamarthi, S., 2005, "Engineering Platforms with Standardization and
Status and Promise," Artificial Intelligence for Engineering Design, Analysis Shared Workplaces," 2005 Innovations in Product Development Conference -
and Manufacturing, 18(1), pp. 3-20. Product Families and Platforms: From Strategic Innovation to
[12] Simpson, T. W., Siddique, Z. and Jiao, J., ed., 2005, Product Platform and Implementation, Cambridge, MA.
Product Family Design: Methods and Applications, Kluwer Academic [34] Nidamarthi, S., 2005, "Architecting and Implementing Profitable Product
Publishers, New York. Families and Shared Engineering Platforms," Product Platform and Product
[13] Pine, B. J., II, 1993, Mass Customization: The New Frontier in Business Family Design: Methods and Applications, T. W. Simpson, Z. Siddique and J.
Competition, Harvard Business School Press, Boston, MA. Jiao, eds., Springer, New York, pp. 475-498.
[14] Meyer, M. H., 2005, "The Power of Product Platforms," 2005 Innovations [35] Berglund, F. and Claesson, A., 2005, "Utilising the Concept of a Design's
in Product Development Conference - Product Families and Platforms: From Bandwidth to Achieve Product Platform Effectiveness," International
Strategic Innovation to Implementation, Cambridge, MA. conference on engineering design, ICED05, Melbourne, Australia.
[15] Mugge, P. C., 2004, "How the NPD Process Changes with Platforms," [36] Claesson, A. and Berglund, F., 2005, "Design Bandwidth," 2005
Platform Management for Continued Growth, Atlanta, GA, IIR/PDMA. Innovations in Product Development Conference - Product Families and
[16] Gerstner, L. V., Jr., 2002, Who Says Elephants Can't Dance? Inside IBM's Platforms: From Strategic Innovation to Implementation, Cambridge, MA.
Historic Turnaround, Harper Collins, New York. [37] de Weck, O. and Suh, E. S., 2006, "Flexible Product Platforms:
[17] Nesland, S., 2004, "Facilitating the Transition from Product to Platform: Framework and Case Study," ASME Design Engineering Technical
The Initial Implementation of a Platform Development Management System," Conferences, Philadelphia, PA, ASME, Paper No. DETC2006/DAC-99163.
Platform Management for Continued Growth, Atlanta, GA, IIR/PDMA. [38] Martin, M. V. and Ishii, K., 2002, "Design for Variety: Developing
[18] McGill, M., 2004, "Aligning a Platform-Based Approach to Product Standardized and Modularized Product Platform Architectures," Research in
Strategy - A Case Study of New Vaccine Development to Licensure," Platform Engineering Design, 13(4), pp. 213-235.
Management for Continued Growth, Atlanta, GA, IIR/PDMA. [39] Seepersad, C. C., Mistree, F. and Allen, J. K., 2002, "A Quantitative
[19] O'Banion, J., 2004, "The C-130 Product Family," Platform Management Approach for Designing Multiple Product Platforms for an Evolving Portfolio
for Continued Growth, Atlanta, GA, IIR/PDMA. of Products," ASME Design Engineering Technical Conferences - Design
[20] Schlueter, J., 2004, "Prepaid Wireless: Launching a New Platform for the Automation Conference, Montreal, Quebec, Canada, ASME, Paper No.
Burgeoning Market of Pay-as-you-go Services," Platform Management for DETC2002/DAC-34096.
Continued Growth, Atlanta, GA, IIR/PDMA. [40] Fellini, R., Kokkolaras, M. and Papalambros, P. Y., 2005, "Commonality
[21] McGrory, P., 2005, "Product Platforms: An Industrial Design Perspective," Decisions in Product Family Design," Product Platform and Product Family
2005 Innovations in Product Development Conference - Product Families and Design: Methods and Applications, T. W. Simpson, Z. Siddique and J. Jiao,
Platforms: From Strategic Innovation to Implementation, Cambridge, MA. eds., Springer, New York, pp. 157-185.
[22] Kaiser, M. L., 2004, "Platform Design for Multi-Faceted Application [41] Simpson, T. W., 2005, "Methods for Optimizing Product Platforms and
Products Using a Customer Driven Product Definition Process," Platform Product Families: Overview and Classification," Product Platform and Product
Management for Continued Growth, Atlanta, GA, IIR/PDMA. Family Design: Methods and Applications, T. W. Simpson, Z. Siddique and J.
[23] Hutchinson, E., 2004, "Developing Organizational Capabilities to Jiao, eds., Springer, New York, pp. 133-156.
Respond to Diverse Product & Technology Platforms," Platform Management [42] Kalligeros, K. C., de Weck, O., de Neufville, R. and Luckins, A., 2006,
for Continued Growth, Atlanta, GA, IIR/PDMA. "Platform Identification using Design Structure Matrices," Sixteenth Annual
[24] Otto, K., 2005, "Model-based Design and Verification of Modular International Symposium of the International Council On Systems
Platforms," 2005 Innovations in Product Development Conference - Product Engineering (INCOSE), Orlando, FL.
Families and Platforms: From Strategic Innovation to Implementation, [43] Oh, J.-S., 2005, "Platforming for Smart Appliances," 2005 Innovations in
Cambridge, MA. Product Development Conference - Product Families and Platforms: From
[25] Kokkolaras, M., Fellini, R., Kim, H. M. and Papalambros, P. Y., 2005, Strategic Innovation to Implementation, Cambridge, MA.
"Analytical Target Cascading in Product Family Design," Product Platform [44] Shooter, S. B., 2005, "Ice Scraper Product Family Development at
and Product Family Design: Methods and Applications, T. W. Simpson, Z. Innovation Factory," Product Platform and Product Family Design: Methods
Siddique and J. Jiao, eds., Springer, New York, pp. 225-240. and Applications, T. W. Simpson, Z. Siddique and J. Jiao, eds., Springer, New
[26] Weiss, K., 2005, "Software Product Line Architecture," 2005 Innovations York, pp. 459-473.
in Product Development Conference - Product Families and Platforms: From [45] Bailey, T., 2005, "Platforming Strategy at Carrier," 2005 Innovations in
Strategic Innovation to Implementation, Cambridge, MA. Product Development Conference - Product Families and Platforms: From
[27] de Weck, O., 2005, "Determining Product Platform Extent," Product Strategic Innovation to Implementation, Cambridge, MA.
Platform and Product Family Design: Methods and Applications, T. W. [46] Otto, K. and Hölttä-Otto, K., 2005, "Platform Concept Selection," Product
Simpson, Z. Siddique and J. Jiao, eds., Springer, New York, pp. 241-301. Platform and Product Family Design: Methods and Applications, T. W.
[28] Johnson, D. A. K., 2005, "The Power of Markets in Product Platforming," Simpson, Z. Siddique and J. Jiao, eds., Springer, New York, pp. 49-72.
2005 Innovations in Product Development Conference - Product Families and [47] Harris, S. E., 2004, "Faster Time to Market with Better Quality Using
Platforms: From Strategic Innovation to Implementation, Cambridge, MA. Product Line Practices," Platform Management for Continued Growth,
[29] Govindarajan, K., 2005, "Enabling Mass-Customized Services: Towards a Atlanta, GA, IIR/PDMA.
Platform for Service Design and Delivery," 2005 Innovations in Product
Development Conference - Product Families and Platforms: From Strategic
Innovation to Implementation, Cambridge, MA.
[30] Jiao, R., Zhang, L. and Pokharel, S., 2005, "Process Platform and
Production Configuration for Product Families," Product Platform and
Product Family Design: Methods and Applications, T. W. Simpson, Z.
Siddique and J. Jiao, eds., Springer, New York, pp. 377-402.

10 Copyright © 2006 by ASME