In his classic work, The Sciences of the Artificial, H.A. Simon (1996) makes two profound statements. The first is the observation that “the natural sciences are concerned with how things are; design, on the other hand, is concerned with how things ought to be.” The second observation is the assertion that design thinking has a unique characteristic that makes it possible for it to be “universalized.” This means that design could be used as a vehicle to communicate across disciplines.

Any professional whose task is to create, to dissolve problems, to choose, and to synthesize is involved with design thinking. Many beautiful designs have been produced in a variety of fields ranging from physical environments and artifacts to music (composition), philosophy (design of inquiring systems), and political and economic arrangements. Some great thinkers have even taken whole societies as systems to be redesigned.

Nigel Cross (2007), in his beautiful book Designerly Ways of Knowing, makes the following indisputable observation:

Design ability is, in fact, one of the three fundamental dimensions of human intelligence. Design, science, and art form an and not an or relationship to create the incredible human cognitive ability (Figure 7.1):

The unique ability of human beings to create images is what design thinking is all about. In this context the distinct advantage of design thinking is to produce new alternatives. It goes beyond default solutions by looking for new exciting possibilities. It is not about selecting the “best” from the existing set of alternatives. The choices in the existing set usually share one or more properties based on an explicit or implicit set of assumptions or constraints produced by the actors' previous experiences with similar situations. The conventional practice of using advanced analytical tools to help select the best alternative will only result in repeating the same known pattern of behavior, since the underlying assumptions governing the generation of alternatives remain unchallenged.

Design thinking, on the other hand, involves challenging assumptions. It represents a qualitative change that includes the notion of beauty and desirability. In this way design results in identifying new sets of alternatives and objectives, looking for more desirable possibilities for the future. Einstein put it so beautifully when he said, “We cannot solve our problems with the same thinking we used when we created them.”

To design anything requires a core concept and basic understanding of the subject to be designed. To design a car one needs to have a concept of a car and be knowledgeable about the state-of-the-art and the relevant technologies. To design a social system a practical knowledge about the nature of sociocultural systems is required. In this context nothing is more practical than a good operational theory. Without an explicit theory about the nature of the beast and the reasons why it does what it does, one would be condemned to keep repeating the same non-solutions all over again. All actions are preceded by some mental image, or theory, about the reality. Nobody, except perhaps a newborn child, is without a theory, no matter how crude and implicit. Those who claim that they are without one are either expecting others to accept their opinions without questions or are simply unaware that they have it by default. Not to get bogged down in a theoretical maze does not mean to fall into the other extreme of mindless action. As long as the assumptions underlying the action are explicit, there exists the chance to learn from experience and improve the quality of our practice. Part Two of this book covered in some detail the critical assumptions about the nature of sociocultural systems, therefore, it is an integral part of our theory of design.

At the core of the design process is the iterative process of holistic thinking. To design is to create structure, functions, and processes in a given context. The context defined by direct user feedback provides the selection criteria for the design process. The point to emphasize is that we cannot deal with context, function, structure, and process independent from one another. We need to use an iterative process to keep the relationships among them interactive and meaningful.

To start, designers must stay at higher levels of abstraction, not letting the tendency to be lost in detail distract them from the ability to be playful in thinking about the whole.

With the first iteration designers will concentrate on developing the desired specifications of the system to replace the existing order. Start by specifying the function — the outputs or the desired impact designers would like the system to have on the larger system of which it is a part. This means understanding the context, especially the behavior of the stakeholders. Who are they and what is their interest? Which variables do they control and which ones do they influence? Designers then will try to understand and define the interdependencies among the many and often diverse desired specifications. They will find out which initial specifications are compatible and reinforcing, which ones are complementary and counterbalancing, and which ones are in conflict. They will also try to reconceptualize any conflicting requirements to dissolve the differences. In the second iteration, designers will let their imaginations take over to create abstract mental images of possible structures and processes that potentially would produce the desired outputs. Then they need to pause and synthesize the selected images into a conjecture (rough approximation of a cohesive whole that only defines primary functions), basic structure, and the outline of the throughput process.

Please recall that the same iterative holistic process that guided the process of inquiry to define problems is also used to design solutions (Figure 7.2).

In the third iteration, they will make a symbolic model of the design to be used to communicate with the design itself and with the stakeholders to achieve consensus that satisfies all concerns. The next iteration will convert this initial rough design into the desired next generation of the system. In successive iterations, after testing for operational viability, more detail and specificity are incorporated into the design. No design will be viable without a thorough understanding of its dynamic behavior. Understanding unintended consequences of our design and ability to map the behavior of positive and negative feedback loops to capture the control system makes operational thinking a critical component of our design theory.

Therefore, holistic thinking, operational thinking, and sociocultural thinking together with design thinking form an interactive set (Figure 7.3). They complement each other to create a competent and exciting methodology that goes a long way in dealing with the emerging challenges of what may seem to be overwhelmingly complex and chaotic social systems.

Design is an intuitive process of creating exciting feasible wholes from messy infeasible parts. It is about selecting a desired future and inventing the ways to bring it about. Design develops innate abilities in dealing with real-world, ill-defined, ill-structured, or “wicked” problems.

Here are ten operating principles of design thinking:

It would be timely here to recall the principle of multidimensionality and the discussion that integration and differentiation form a complementary pair. Their interdependency is such that for any given level of differentiation there is a corresponding level of integration; otherwise, the system will fall into a state of chaotic complexity.

Unfortunately, the increasing amount of subspecialization in every discipline has reached such a level that we have produced an aggregate of subcultures that are not even capable of communicating within the same discipline let alone across disciplines. As a modern scientific culture, we have differentiated without paying any attention to the need for integration. It is no wonder we have lost the ability to connect the dots in a host of many arenas. Despite all claims to the contrary, most contemporary organizations are merely collections of silos of differentiated functional departments — each with very high levels of expertise — without any real means of integration among them. We have naively assumed that coordination is the same as integration. We have erroneously reasoned that if several functions report to a coordinating boss he/she will be able to integrate them into the emerging whole and properly connect the dots. The recent financial crisis, housing bubble, and the obvious messes we are facing in education, immigration, and health-care systems, just to name a few, should provide enough evidence to convince us that our integrative efforts have been a failure.

The failure to integrate specialized parts of the organization makes the following discussion of modular design especially important. Modularity is both a great vehicle for designing complex systems and a perfect mechanism for functional integration at operational levels as well as an effective format to create dynamic structure at architectural levels of social systems.

IBM introduced its formidable architecture for the design of its 360 line of computers in the mid-1960s. This was the turning point that led the way for the creation of the present monumental state of information technology. The design contained a revolutionary conception known as modularity, which transformed the field of complex system design to an unbelievable state of possibilities and potentiality.

Being at the right place and at the right time, I was lucky to be among the first few in the IBM World Trade Corporation who got the call to learn “360 architecture” and the magic of its modular design. The experience had such a profound influence on my professional life that many years later it became the basis for my attempt to develop a systems theory of organization (Gharajedaghi, 1985) and to bring the beauty and immense potentiality of the multidimensional modular design into the social context, more specifically, design of business architecture.

A module is an element in the architecture that is designed in such a way that it has strong internal interdependency (full functional integration) and weak or a minimum external dependencies and interactions. This fully integrated module is then connected to a specific platform designed to simplify the communications or interactions between the modules, its host platform, and the rest of the system. A module, therefore, is an integrative vehicle to facilitate and manage all of the internal interactions among the differentiated functions to produce an emergent whole (a signal or an outcome) to be transmitted or transferred to external elements of the system of which the module is a part.

A platform is a predefined host for a given class of modules. Each module of this class might have its own unique functionality, but all would share one common characteristic that defines the nature of their relationships with the rest of the system's platforms and modules. The platform, in essence, is a predesigned interface with the critical function of simplifying and managing all of the interactions of its class of modules with the rest of the architecture.

Multidimensional modular design is the architecture of a dynamic structure. It is an integrated system of differentiated platforms (each with its own class of modules) with a dynamic but simplified, predefined relationship. In this architecture combinations of different modules from different platforms transform the system into a different design as necessary. For example, we find a telephone, a music box, a GPS, a camera, or hundreds of amazing applications in Steve Jobs' incredible iPhone.

Figure 7.4 is the representation of a multidimensional modular design with the required minimum of three dimensions of inputs, outputs, and markets platforms. The input platform is the host of the class of modules that represents internal resources and competencies of the system such as a knowledge bank, manufacturing unit, and shared services. Input modules are also responsible for continuously refurbishing internal resources and competencies by continuous re-education and research. In addition to mastering their functional expertise, members of the knowledge bank will be required to learn design methodology. On the other side we have a market platform that includes the class of modules with the capability to access different markets and ability to assess their emerging needs and opportunities. Finally, the outputs platform hosts a group of dynamic modules that have the ultimate authority and responsibility for designing products, programs, and projects by continuously matching external opportunities with internal competencies. The products, programs, and projects have varying life spans that range from days to months or even years if necessary. Although output modules control the flow of money they do not own any resources (all capital assets belong to input modules); therefore, resistance to change is minimized.

The following example should help explain the intricacies of modular design and answer questions regarding its setup. A client of mine was in the business of designing custom-made separation units to purify water for customers in the electronics industry. The company's core was its engineering division, which was headed by the chief engineer. Reporting to the chief engineer were the mechanical engineering, electrical engineering, chemical engineering, and process engineering departments. Each of the four engineering departments in turn had its own chief engineer as well as a deputy chief, a senior engineer, and a few junior engineers who actually did the work. The chief engineer was the first to receive a client order secured by the company's sales department, and once the chief approved the order, the design process began. The process involved all four departments but in a successive order. It began by transferring the order to the chief chemical engineer, who then forwarded it to the deputy chief chemical engineer, and then down to the senior engineer who finally selected one of the junior chemical engineers to initiate the design. The junior chemical engineer, upon completing his part of the design, began the ascending order of forwarding his work back up the chain to the senior engineer, followed by the deputy chief, and finally the chief chemical engineer, who then transferred the approved design to the chief process engineer. The whole pattern was then repeated in the process, electrical, and mechanical engineering departments. When the final design was completed, it was then returned to the accounting and sales group for pricing and cost calculations. Once the final budget was priced and shared with the client, the back and forth process of price negotiations between the client and the company's sales and engineering teams began.

This process usually took about six to eight months to complete at an estimated cost of around $500,000. Over time, both the timetable and the associated costs surrounding each bid increased at an annual rate of 5% and this was becoming a major source of friction between the company and its customers.

After reviewing the situation it was obvious that we needed a modular design. But the traditional culture with its strong functional structure was not ready for the kind of change that modular design requires. To reduce resistance to change, we proposed a simple experiment to see if there were other alternative processes the company could consider to reduce cost and improve timing. The experiment was to create a learning cell involving four junior engineers, one from each of the engineering divisions, in addition to a cost accountant and a sales engineer from the sales department. Each member of the learning cell was asked to teach the remaining members of the team the essence and critical aspects of the functions he/she performed, and more specifically how each one of the defining variables impacted the final result. The learning cell was to manage this task on their own time in an effort to prevent this exercise from affecting their present work. As an incentive, each member of the team was promised a $10,000 salary increase if all of the members of the learning cell demonstrated that they had learned all of the functions and critical concerns of this design process from one another and successfully finished an assigned project.

After three months, the group was ready to accept their first project. When they received their first project, all six members of the module met with the client together to learn about its operations, major concerns, objectives, and requirements firsthand and as a team.

It took the modular design team only four weeks after this visit to produce the first iteration of their design. This design, with small modification, was approved by the client and finalized in another week. Total cost was $120,000 and total time was six weeks. This experiment was duplicated twice with similar results, and as such paved the way for the entire organization to convert to a modular design.

We have said that self-organizing, purposeful, sociocultural systems are self-evolving. They do not simply adapt to their environments but co-evolve with them. They can change the rules of interaction as they evolve over time. However, like all living systems, they exhibit a tendency toward a predefined order. Their behavior is guided by an implicit, shared image. They tend to approximate and reproduce a given pattern of existence very closely. To change this pattern of behavior the implicit shared image (the organizing attractor) needs to be changed.

The shared image, itself a complex design, stands at the center of the process of change. Once formed it acts as a filter so that success of any action intended to produce change invariably depends on the degree it penetrates and modifies this shared image.

Ackoff's systems methodology aims at the core of this conception. Its ultimate objective is to replace the distorted “shared image” responsible for regenerating a pattern of malfunctioning order with a shared image of a more desirable future. Working together for decades in INTERACT, the Institute for Interactive Management, we have used the interactive design process to redesign distorted shared images and to produce a lasting change in the behavioral pattern of many social systems.

Designers replace the existing order by operationalizing their most exciting image of the future — the design of the next generation of the system. An explicit and exciting design/image of the future is a powerful instrument of change. The desired future is then realized by successive approximation (inventing the ways to close the gap between actual and desired states).

This pretentious and daring optimism, however, is based on the following assumptions:

Donella Meadows' (2008) insightful observation that “The future can't be predicted, but it can be envisioned and brought into being. Social systems cannot be controlled but can be designed and redesigned” is a welcome confirmation of design thinking.

Interactive design is identified with Russell Ackoff. It is the core of his famous purposeful systems methodology. The ultimate aim of interactive design is to define problems and design solutions in an environment characterized by chaos and complexity. According to Ackoff (1981), “We fail more often not because we fail to solve the problems we face but because we fail to face the right problem.” We have routinely been taught how to solve problems, but rarely how to define one.

Traditionally, there are three ways to define a problem. The most common approach defines problems as deviations from the norm. The major shortcoming of this approach, besides the difficulty in defining the “norm” in a sociocultural system, is to reinforce the existing order. This is usually done despite strong suspicion that the existing order might be the source of the problem. A simple example is the “back to basics” movement in education.

A lack of resources is the next popular way to define a problem. It seems that somehow we cannot get enough information or money, and most certainly we do not have enough time to deal with most situations. This should hardly be a surprise, since time, information, and money are universal constraints. We will never have enough money. We will never have enough time. We will never have enough information. The more we know, the more we know that we do not know. A minister of economy in my native country once asked me to help him assess the impact of a certain decision on three important factors he was concerned with. I told him it would take me a month to develop the proper model. He replied, “The decision is going to be made without you. If you want to have any influence on this one, be in my office with your model at 7:00 a.m. Monday morning. Otherwise, get the hell out of the way.”

The third, and perhaps the most obstructive, way to define a problem is the tendency to define it in terms of the solution we already have. Existing solutions conveniently shield us from seeing the reality, so we accept the problem at face value. Not surprisingly, an operations researcher may see a situation as an allocation problem, while an accountant may consider it a cash flow problem.

Those of us trained as professionals — engineers, doctors, and lawyers — come with a tool bag. In each area we have been exposed to a series of classic cases, which supposedly resemble the problem set we are expected to encounter in our professional lives. We learn the solutions to these problems at the professional schools and store them in our tool bag for future use. What we have to do in real-life situations is find similarities between the situation we face and one of those cases, and then simply apply the proposed solution in the case to the problem at hand. This approach is so ingrained in the way we do things that we are not even willing to entertain a question if we do not already have the answer. A valued client once protested, “Why are you pressuring me to face this problem when you know very well that I don't have a solution for it?” It was not easy to convince him that today's problems no longer yield ready-made solutions and that his job had changed from a tool user to a tool maker. I had to remind him that he was paying me quite handsomely to help him do so. Unfortunately, even if we have a potent and innovative solution at hand, most of the time we lack the confidence to use it. We need to know who has done it before. It is as if, despite all claims to the contrary, we do not dare to be the first. Stafford Beer (1967) has expressed this phenomenon elegantly: “Acceptable ideas are competent no more, but competent ideas are not yet acceptable. This is a dilemma of our time.”

We have said that no problems or solutions are valid free of context. A phenomenon that can be a problem in one context may not be one in another. Likewise, a solution that may prove effective in a given context may not work in another. However, the tendency to define the problem in terms of the solution, and a strong preference for the context-free solution that is tried and true, create a closed loop. The process keeps on regenerating the same pattern of behavior all over again.

In a chaotic and complex environment we are not confronted with a single problem but a system of interdependent problems, or what Ackoff calls a mess. A mess is neither an aberration nor a prediction. It is the future implicit in the present operations, pointing to the potential seed of its destruction.

To map the mess is to capture the iterative nature and the workings of the multiple feedback loops that form critical interdependencies and define the second-order machine responsible for regenerating the problematic patterns repeatedly. Perhaps the best example of mess formulation is Das Capital. Ironically, the most important contribution of Karl Marx was not the solution he proposed but the problem he defined.

A good formulation of the mess makes a convincing case for fundamental change and sets the stage for effective redesign. A detailed discussion of how to formulate a mess will come in Chapter 8. The rest of this chapter deals with designing solutions, which consists of two distinct phases, idealization and realization.