Accelerating Manufacturing Innovation Through Collaboration

Over the past decade, the manufacturing sector has been under enormous and constant pressure to not only maintain, but to increase its competitiveness.


Over the past decade, the manufacturing sector has been under enormous and constant pressure to not only maintain, but to increase its competitiveness.  Prompted by macroeconomic trends such as global expansion, more virtualized business models, sharp rises in energy costs and the need for collaborative process integration, enterprises have been able to radically transform the performance of their value chains.  However, for enterprises who want to consistently excel in such a challenging environment, excellence with innovation has become a mission imperative.

Innovation is the central issue in economic prosperity.[1] 

Michael E. Porter

Within the manufacturing sector, innovation is a term that generally refers to technological advancement such as new products and features, materials fabrication processes and equipment, green and/or more environmentally friendly processes, end products, etc.  However, this paper proposes that the value of innovation is much broader than merely technological and procedural. 

BT’s point of view is that successful innovation addresses the broader strategic dimensions of enterprises’ strategies that focus on objectives such as branding image, desire to become more customer focused, compressing concept-to-market cycle time, success with exploiting and creating more collaborative value chains, information availability and knowledge sharing that yields value.  Critical to implementing such strategies, effective collaboration must become a core competency of the enterprise and its partners.

The Innovation Paradox

In the past, internal R&D was universally viewed as a strategic asset, as well as a barrier to competitive entry into most industries.  Only large and financially strong enterprises with significant staff resources and long-term research programs could compete. Research-based companies such as Westinghouse, IBM, GE, Pfizer and AT&T conducted the majority of research in their respective industries, and in turn also garnered most of the related revenues from their discoveries. However, these same former leading industrial enterprises are finding remarkably strong competition from numerous newer companies such as Intel, Google, Genentech, and many others who conduct comparatively little primary research on their own.  Although these upstarts have been very innovative, they have pursued a strategy of innovating with the research discoveries of others.

No longer is the creation of new ideas and techniques an advantage limited solely to large centralized R&D departments.  Much to the contrary, innovations are increasingly brought to the market by collaborative networks of enterprises and individuals, selected according to their comparative advantages.  In this new business model, enterprises decompose the innovation value chain and source elements to partners that possess critical advantages that include specialized skills, agility and flexibility, lower cost structures, access to emerging markets and many other characteristics that can provide a source of tactical differentiation.  The aim of these emerging business models is to establish mutually beneficial relationships through which new products and services can be developed and brought to market.  In simple terms, enterprises are increasingly seeking superior performance in innovation through strategies of partnering and collaboration. 

This new business model is being driven by a series of trends forcing enterprises to re-examine traditional approaches to achieving innovation.  First, the complexity of products and services has and continues to increase at an ever-accelerating rate, both in terms of the number of technologies incorporated into the finished good, as well as suppliers thereof.  Consider the average modern automobile, with literally thousands of physical parts, software, microchips, sensors, transmitters, drive-train components, etc. from suppliers based all over the world.  No longer is it possible for a single company to master all the prerequisite skills and co-locate them within a single facility. Secondly, a supply of skilled low cost labor has emerged in developing countries, creating incentives to substitute these resources for existing higher-cost equivalents.  Thirdly, different regions of the world have developed unique skills and capabilities, which most manufacturers have already begun exploiting for multiple advantages.  Finally, advances in design tools and technology combined with the rise of more easily integrated collaborative software architectures and data exchange standards have in combination driven down the costs associated with coordinating distributed work.  Quite simply, collaborative design and manufacturing are no longer leading edge theories or visions, but have become commonplace across most areas of the manufacturing industry.   That’s why we can have high-quality and reliable automobiles, with design, assembly and component parts from literally hundreds of globally spread suppliers.

In the opening sentence of Charles Dickens’ “A Tale of Two Cities” the reference is made to the time of the French Revolution as “it was the best of times, it was the worst of times.”[2]  This observation could easily be applied to the current state of manufacturing innovation.  Mankind’s seeming comprehension and mastery of the engineering of materials, biological evolution of species and laws of physics have accelerated at a pace whereby modern society is rarely surprised by most new innovations.  Even in the most ancient of industries such as retailing, enterprises are continually innovating by employing software and network-based technologies to create pull-driven supply chains, whereby physical products are being increasingly replaced with information (e.g. Kanban replenishment triggers), thereby enabling retailers and manufacturers to shrink buffer inventories, reduce stock-outs, while meeting customers’ expectations for immediate product availability.

Yet in many ways, it is the worst of times for enterprises that have historically based their branding and profitability on constant streams of new product innovation.  To illustrate, consider the premier industrial research laboratory of the last century, Bell Labs.  At that time, Bell Labs represented an unquestionable strategic advantage in Lucent’s (now part of Alcatel-Lucent) competition against Cisco in the telecommunications hardware market.  Despite Lucent’s advantages in many critical areas, including R&D, Cisco consistently grew market share against Lucent. Cisco seemed to introduce many new products and services, despite its lack of comparable research capabilities.  Though they were direct competitors in a very technologically complex and fast paced industry, Lucent and Cisco were not accomplishing innovation using the same strategies.  While Lucent committed enormous economic and personnel resources to exploring new materials and state-of-the-art components, Cisco conducted very little internal primary research, choosing instead to pursue a much different strategy in its struggle for innovation leadership.  Cisco’s strategy was to scan the world of small start-up companies that were springing up all around it and that were commercializing new products and services, which the market would ultimately judge to be successful or not.  In executing this strategy Cisco would occasionally invest in these start-ups by providing needed capitalization and other times it would simply partner with them.  In many cases, Cisco ultimately acquired the startups and assimilated them into their growing enterprise.  This strategy permitted Cisco to keep up with the innovation output of its industry, without performing significant internal R&D or committing substantial capital to new technologies, the majority of which ultimately failed to survive in the marketplace.

The outcome of these contrasting strategies can be best measured by Alcatel-Lucent’s February 8, 2008 posting of a $3.74 billion fourth-quarter 2007 loss and cancellation its 2007 dividend, after taking a write-down of more than $3 billion on its U.S. wireless business.[3]  In contrast, Cisco Systems reported fiscal year 2007 net sales of $34.9 billion (an increase of 23% over the prior year) and an annual 2007 net income of $7.3 billion.[4]

Lucent’s experience with the limits of its internally-focused innovation strategy is not at all unique.  For example, IBM’s research prowess in mainframe-based computing and operating systems were of little advantage against Intel and Microsoft in the personal computer business.  These examples lead to a fundamental paradox that confronts virtually all innovating enterprises.   That being, while there is no dearth of good ideas generated within high-powered and talent-rich R&D centers, internal industrial research is often far less effective at generating innovation output to the marketplace. 

Changing Innovation Paradigm

So, what are the forces that have made it so difficult for leading enterprises to sustain their innovation leadership? 

Within the last decade we are witness to a paradigm shift in how enterprises approach and conduct innovation.  The old paradigm is often referred to as a “Closed Innovation Model”[5], which is based upon a philosophy under which successful innovation is governed by command and control.  In a closed innovation model, enterprises must generate most of their own ideas, develop them, as well as build, finance, distribute, service and evolve them within the enterprise to ultimately reach the market.  Central to this strategy is an over-arching philosophy that the enterprise must be self-reliant, essentially because of a lack in faith in the quality, availability, capabilities, reliability and intentions of others.  This logic is the underpinning driver of closed innovation.

In a closed innovation environment, ideas flow through the enterprise from R&D to the market.  Ideas are screened and refined during the R&D process, and those surviving ultimately emerge into the market place, which judges financial viability.  This linear and sequential approach to operating concept-to-market processes is designed to refine innovation alternatives, with those surviving having a greater chance of success potential in the market because their progression has been better managed and controlled.

In the past decade several factors have combined to erode the underpinnings of closed innovation strategies.  One factor has been the growing mobility of highly experienced and skilled personnel.  When people left a company after working there for in some cases multiple decades, they took a good deal of that valuable knowledge with them to their new employer.  A related erosion factor was the amount of company-funded college and post-graduate training that many people obtained.  Also, the breakup of the former Soviet Union made many highly-educated and skilled scientists and engineers available to private industry on a global basis.   Ultimately, this growing number of all such people allowed the capabilities to perform innovation to spill out of the knowledge silos of corporate research labs to companies of all sizes in many industries and geographies.  Finally, a further factor was the growing presence of private venture capital, which specialized in creating new companies that commercialized external research and converting these start-ups into fast growth, highly-valuable companies. Often, these highly capable start-up companies became formidable competitors for the large, established firms that had formerly financed most of the R&D in their industries.

The logic of closed innovation was further challenged by the increasingly rapid cycle-time of concept-to-market for many products and services, making the shelf-life of virtually every technology ever shorter.  Moreover, increasingly knowledgeable customers and suppliers challenged the innovating companies’ ability to profit from their own knowledge silos (i.e. through multi-stage competitive bidding, commoditization of components, etc.)  Once these erosion factors have cumulatively impacted an industry, the assumptions and logic that once made closed innovation an effective approach no longer applied.  The bottom line is simply that in situations where these erosion factors have taken root, closed innovation is no longer a profitable or sustainable strategy. 

For these situations, a new approach, often called “Open Innovation”[6] has emerged.  The Open Innovation Model is a strategy that assumes enterprises can use both internal and external ideas, as well as internal and external paths to markets, which can (and should) include the creation of new markets.  Open innovation combines internal and external ideas into innovations whose requirements are defined by a business model that utilizes both internal and external ideas to create value, while defining internal controls to ensure some portion of that value is captured.  Open innovation assumes that internal ideas can also be taken to market through external channels, outside the current businesses of the enterprise, to generate additional value. 

An excellent example of the relevance of open innovation comes from Procter & Gamble, who in the late 1990s decided to change its approach to innovation.  The firm extended its internal R&D to the outside world through an initiative called “Connect and Develop.”  This initiative emphasized the need for P&G to reach out to external parties for innovation ideas.  The company’s rationale was simple – Inside P&G there are roughly 8,000+ scientists advancing the industrial knowledge that enables new P&G offerings, but outside there are 1.5 million!  So why try to invent everything internally?  “P&G’s strategy of open innovation now produces more than 35% of the company’s innovations and billions of dollars in revenue.”[7]

A survey conducted by BT of several dozen leading manufacturing and Fast Moving Consumer Goods (FMCG) enterprises found that while reducing R&D costs is often the number one priority for enterprises using partners to innovate, these savings are often lower than expected, due principally to the added costs associated with the need for greater coordination.   In this survey group, enterprises most successful with open innovation models typically focused greater attention on how to leverage partner capabilities, than merely focusing on partners as merely a means to achieve lower costs through wage arbitrage.  In this survey, BT observed two broad types of capability in action:  First, the ability to rapidly bring online large amounts of research capacity, allowing enterprises to shrink concept-to-market and increase responsiveness, while avoiding start-up capital investments and the cost of full-time staff; and second, the ability to access unique competencies, technical know-how and/or process expertise that enterprises simply did not possess internally.  Successful enterprises sought partners with a blend of both capabilities, giving them almost instant access to a repertoire of capacities and competencies not available in-house.  As one manager recalled, “It usually takes us nine months to find and hire a single new materials testing engineer, but using our partners, we staffed up the full project team in only two weeks, accessing critical skills that we didn’t have internally.”

BT’s survey also observed that leading enterprises viewed partners as an extension of their own development organizations, seeking their participation in meetings and even including them in internal communications.   As part of this philosophy, they required greater continuity in partner staffs, in contrast to a transactional model, in which people move in and out of projects.  This ensured the “tacit” knowledge of a project’s context was retained, and improved communication between teams.  As one manager explained, “It takes time to appreciate the skills of each team member and understand how to work together.  When people leave, we have to go through that learning curve again.  So we put a premium on ensuring staff continuity. 

The final area separating leading enterprises from others was their willingness to invest in developing “collaborative capabilities.”  All too often, enterprises assumed that their existing staffs, processes and infrastructure were capable of meeting the challenge of collaboration.  But successful collaboration doesn’t just happen – it is a capability that must be learned.  Rarely do enterprises get it “right first time.”  Leading enterprises recognized this reality, and made investments to enhance their collaborative performance over time.

Successful enterprises targeted investments in four principal areas: people, process, platforms and programs. These investments were typically funded outside the budgets of individual projects, given few projects can justify the levels of infrastructure needed to perform well on their own.  In essence, leading enterprises made a strategic decision to invest in collaborative capabilities, and sought to leverage these investments across projects and over time.

BT’s survey found that leading enterprises developed technology “platforms” to improve the coordination of work. These platforms comprised; (a) Collaboration tools and technologies to improve the effectiveness of distributed work, (b) Technical standards and interfaces to ensure the seamless and secure integration of partner inputs and outputs, (c) Rules to govern the sharing of intellectual property among partners and (d) knowledge management systems to capture the enterprise’s experience on how distributed work is best performed.  This collaboration “infrastructure” was then leveraged across multiple projects over time, with the goal being to promote a long-term view of continuous improvement in effective collaboration.

The importance of each of the four areas (people, process, platforms and programs), as well as their interdependencies cannot be understated in terms of driving end results or outcomes.

To illustrate, consider the recent troubles at Airbus, in developing its flagship A380 aircraft.  Airbus’ German and French partners chose to work with different versions of Dassualt Systems’ CATIA design software.  However, design information in the older system was not translated accurately into the new one, which held the “master” version.  Without a physical mock-up, these problems remained hidden throughout the project.  The result:  300 miles of wiring, 100,000 wires and 40,000 connectors that did not fit, leading to a 2-year production delay at a cost of $6bn.[8] Yet, the cause of Airbus’s problems was not in choosing different software versions; rather it lay in the lack of an effective process for dealing with the problems this inconsistency created, and means of identifying and working around such problems.

A striking contrast of these dynamics is seen in Boeing’s development of its 787 “Dreamliner” aircraft.  Boeing builds among the most complex commercial products in the world, each project being almost literally a “bet-the-company” experience.  The levels of capital investment required and the increasing breadth of technologies that must be mastered, from digital cockpit design to new lightweight materials, have forced Boeing to look at new forms of innovation, the aim being to share risk with partners while exploiting the unique technical expertise that each brings to product design and development.

Boeing’s approach to the 787 was the epitome of global collaboration and concurrent engineering. The project included over 50 partners from over 130 locations working together for more than four years.  From the start, the aim was to leverage advanced capabilities from this collaboration network.  For example, in technologies like composite materials, which are being used for the first time for large sections of the airplane, smaller more focused firms had developed expertise that was unique.  Rather than replicate this expertise, the firm sought to tap into it, blending it with skills from other partners developing complementary technologies.  Furthermore, the relationships it established were not the traditional “build-to-print” contracts of past years.  Instead, partners designed the components they were to make, ensuring a seamless integration with the outputs of other partners.

Boeing thereby represents a very successful example of a strategic migration to an open innovation model.  That is, Boeing’s source of competitive advantage is increasingly less related to the possession of deep individual technical skills in hundreds of diverse disciplines.  While the firm still possesses such knowledge, this is no longer what differentiates it from competitors such as Airbus, who can also access similar capabilities. Rather, Boeing’s unique assets and skills are increasingly tied to the way the firm orchestrates, manages and coordinates its network of hundreds of global partners and executes concurrent engineering processes.  Boeing’s experience is increasingly common across the industries we observed in the survey, with collaboration becoming a new and important source of competitive advantage.

Concurrent Engineering

The exchange of information (collaboration) is the lifeblood of innovation and product development.   When an electronics company’s integrated circuit designers can collaborate with the cabinet design engineers, they can design a better-fitting circuit for the enclosure and address other specifications dealing with heat, Electromagnetic Field (EMF) considerations, etc. Similarly, when the enclosure designers know precisely what the circuit designers’ specifications are, they can design a cabinet where it’s easier to install the circuit components.  Such collaborative flows of information allow for experimentation and innovation, and for that reason, many leading enterprises, such as Boeing, encourage collaboration in their product development processes.  This iterative and collaborative process is known as concurrent engineering.

All products, as well as their fabrication techniques, have a need to incorporate constraints imposed by the various manufacturing processes in the overall finished product design.   Addressing these design constraints early in the development process creates the opportunity to reduce manufacturing costs and ultimately improve product quality.  Often the method of accomplishing the integration of design with other functions is through the use of cross-functional and cross-enterprise teams. These teams may include individuals with expertise in marketing, product design, tooling, quality control, patient protection, finance and numerous others depending on the type of product.

Concurrent engineering is a business strategy which replaces the traditional product development process with one in which tasks are performed in parallel and there is an early consideration for every aspect of a product’s concept-to-market process.  This open innovation strategy focuses on the optimization and distribution of an enterprise’s resources (both internal and external) in the design and development processes to ensure effective and efficient product delivery.

Similar to open innovation, concurrent engineering is increasingly recognized as a strategic weapon that businesses use for effective and efficient product development.  It is not a trivial task, but a complex strategic plan that demands full corporate commitment, therefore strong leadership and teamwork go hand and hand with successful concurrent engineering programs.  Also, as with open innovation it requires the collaboration tools and technologies to improve the efficiency of distributed work.

The lesson is clear, innovation must be carefully planned and managed, and enabled by effective collaborative processes and supporting tools.  To do that, management needs representation tools to aid in analyzing information flows, which represent the collaboration between work groups. 

For product development the complexity of the projects often exceed the analytical capability of any single available tool or perspective.   We are reminded of the story about the blind men touching various parts of an elephant and trying to make inferences as to what the thing might be.  One touches a foot; another, a tusk; another a trunk.  Each one, isolated from the others, is mystified about the identity of the whole, until they confer over how the separate pieces produce a solution to the puzzle.

There is, however, an analytical tool that can be used to obtain a simple and meaningful representation of collaboration processes. This tool, the Design Structure Matrix (DSM), focuses on representing the information flows and thereby greatly aids in establishing the collaboration requirements that support innovation and product development. As a result, it is better able to depict the key dynamics and relationships of innovation processes.  What’s even more useful is that it can often provide representations of complex development processes on a single page, versus volumes of flow charts and Entity Relationship Diagrams (ERDs).

Design Structure Matrix

The Design Structure Matrix or DSM is an analytical tool, which can be used to map information flows and the related collaborations necessary to support the product development processes.  DSM represents visually the network of interactions among development participants and facilitates analysis of the collaboration implications.   Hence, DSM can serve as a vital tool in implementing open innovation methodologies and the supporting technology platforms.

The key differentiator (and virtue) of the DSM over other conventional analytical tools in that it focuses on representing collaborative information flows rather than work flows.  Thus, it is better able to depict the key dynamic of the innovation process – COLLABORATION.

The analytical process for creating and using the DSM is very straightforward. Rows and columns are headed with the complete list of functional groups and the related activities to be performed in the project.  Marks (represented by “X’s”) in the matrix are used to designate if there are information-based collaborations among the groups/activities.  Marks in a single row (of the DSM) represent all of the groups whose output is required to perform the task corresponding to that row.  Similarly, reading down a specific column reveals which group receives information from the task corresponding to that column.  If the order of elements in the matrix depict a time sequence, then marks below the diagonal represent forward information transfer to later (i.e. downstream