Editor | On 13, Feb 2004

By: Tilo Pfeifer, Martin Tillmann

Abstract
The enormous potential of process innovation and optimization is often undervalued. In order to stay competitive organizations are forced to adapt quality of products and production to permanently growing requirements. Modern manufacturing processes need to fulfill highest demands on efficiency, flexibility and reliability. Thus, process optimization gains significantly in importance. The increasing complexity and integration of production systems complicate their optimization. Hence, the classical approaches of manufacturing process optimization reach their limits. Depending on the development state of the production system, exponential improvement can only be achieved, using a form of optimization that considers and questions the existing system structure as a whole and enables innovative process design. However, up to now operational practice mostly focuses on the improvement of single process steps and optimizes only within the existing structures. Thus, a large potential for optimization stays unutilized in the manufacturing area.

The objective of the project »Innovative Optimization of Process Chains« which was conducted by the Fraunhofer IPT was to develop a system for the innovative optimization of process chains in manufacturing. The system enables a holistic consideration and optimization of manufacturing processes that results in an innovative process design. To achieve this goal a procedure was developed that integrates classical tools of quality management as well as innovative methods for manufacturing optimization. These are among other things the tools of the so-called »Theory of Inventive Problem Solving (TRIZ/TIPS)« and the »Theory of Constraints (TOC)« which promise a high profit and a relative low effort for qualification. TRIZ and TOC share the assumption that most core problems exist because some underlying conflict or contradiction prevents the straightforward solution of a problem (e.g. improving one characteristic leads to impairment of another). TRIZ especially provides tools to find breakthrough solutions (innovative process design) that overcome these conflicting objectives. TOC is a systems approach to continuous improvement of process chains. It provides a set of five powerful tools, three of them focusing on the safe implementation of solutions found for the problems of optimization, particularly since most great ideas fail in the implementation stage.

The research project was accompanied by a workgroup that consists of representatives from different industries. The suitability of the systematic for industrial practice was validated in the course of two pilot-applications. The presentation will provide an overview of the research project and give an introduction into the developed system and it’s tools. The application of the systematic and the DMAIC cycle will be explained on basis of examples from the pilot applications.

Challenges of Process Optimization in Manufacturing
The competitivenes s of modern-day companies is more than ever a direct function of their ability to implement process innovations and optimization. Such optimization, however, is becoming increasingly harder to realize, due to the growing complexity of the ways in which the networks of structures and processes in the manufacturing area are interconnected.

In operational practice, process optimization often means trouble-shooting. People in charge of (continuously) improving production processes strive towards “problem solution in a single stroke”. The analysis of the problem frequently gets a raw deal. If any, primarily basic tools for problem solving (e.g. brainstorming) and trial-and-error are applied. The quality of the solution is strongly dependent on the experience and intuition of the problem-solver. Furthermore, many great ideas fail in the implementation stage. When the complexity of the problem is high and no obvious solution is known this approach reaches its limits /Lit 1/.

Quality management in manufacturing provides several tools to support optimization efforts. FMEA, SPC or test data evaluation can be used to reveal problem areas in a process chain. DoE is a powerful tool that helps to understand an individual process and to simultaneously optimize it for all critical outputs. However, classical QM-tools for manufacturing are mainly focused on the prevention of failures without considering and questioning the existing structures as a whole. The efficiency and the effectiveness of a process chain are not only measures of defects but are multidimensional. This is where classical methods often run into their limits. Technologically mature systems – which serial production facilities invariably are – can only be improved on a profound level by integrated system analyses and holistic optimization efforts that may result in innovative process design. This means that a large part of the optimization potential in the production sector and neighboring fields remains untapped so far. A holistic approach for innovative process chain optimization is required. This approach needs to utilize classical and innovative methods that can cover the integrated analysis of the process chain, the systematical generation of innovative solutions and their safe implementation.

The Project »Innovative Process Chain Optimization (IPO)«
The objective of the research project (duration: Dec. 1st 2000 until Dec. 31st 2002) was the development of a system which allows the holistic optimization of process chains in the manufacturing area with a view to increasing quality and performance (see slide 4). For practical purposes, a pertinent procedure was developed which successfully combines conventional quality management methods with new and integrated approaches. These incorporate the Theory of Inventive Problem Solving (TRIZ/TIPS) and the Theory of Constraints (TOC), both of which provide high-performance tools for process optimization. The palpable results of the project include an advanced tool kit for quality management operations in manufacturing business environments. The contained tools support the individual stages of the optimization process to which they can be clearly and easily assigned.

The internal experts are supported with learning and applying both the procedure and the integrated tools by a guidebook and a corresponding qualification module. The qualification module and the guidebook are now available as a CD-ROM (presently only in German). The CD-ROM also includes the IPO-tools with step-by-step instructions for their application as well as templates and worksheets. The CD-ROM can be found with the final report (DGQ-FQSBand 86-03) of the IPO-project, that can be ordered from the Federation for Quality Research and Science (FQS).

The project »Innovative Optimization of Process Chains« was sponsored by the German Federation of Industrial Cooperative Research Associations (AiF) and was funded by the Federal Ministry of Economics and Technology (BMWi). It was administrated by the Federation for Quality Research and Science (FQS). To ensure the suitability for industrial practice, the project was supervised by a working group that consisted of companies from various industries. The developed system (procedure and toolset) has been successfully validated in the course of two pilot applications in different companies.

TRIZ – Systematically to Innovative Solutions
In the optimization of a process chain the project team has to face a problem which is usually characterized by many requirements and objectives, some of which are conflicting. The team may have to solve a problem with no known solution. This is called an inventive problem and may contain contradictory requirements. Knowledge and creativity are two essential conditions for a successful solution. However, there is often a lack of both /Lit 2/.

Even though the composition of the team is interdisciplinary, it is virtually impossible to integrate universal knowledge of all specialized areas into a team. Independent studies have shown that creativity diminishes steadily throughout the work phase of life /Lit 3/. Many people hesitate to be creative, because they fear that they lack the essential skills. In general humans solve problems by analogical thinking. That is, we try to relate the problem we are facing to some standard class of problems (analogs) we are familiar with, and for which a known solution exists. If we can draw the right analogy, we can find the right solution. Our knowledge of such analogous problems, however, is the result of our educational, professional, and life experiences. Ideally, all potential directions for solutions should be equally regarded. In reality however, only solutions within one’s own experience are considered while the consideration of alternative technologies to develop new concepts is ignored /Lit 4/. This results in what is called psychological inertia which defeats randomness and leads only into those areas of personal experience.

For process optimization it would be a decisive advantage if the team had an extensive knowledge base and was capable of generating innovative concepts purposefully and systematically, rather than more or less at random. The TRIZ method provides some suitable tools. TRIZ expands the knowledge horizon of the developer by using a scientific-engineering knowledge base and supports the user systematically throughout the process of creative problem solving. The method ensures an effective and efficient search for innovative solutions, focusing on the so-called Ideal Final Result. It limits the search field considerably, but fosters creativity within that search field. TRIZ also helps the user to detach himself from the psychological inertia vector, i.e. from his usual thought patterns and structures /Lit 5/.

TOC – From Concept to Implementation

With the Theory of Constraints (TOC) another powerful methodology for process chain optimization was identified. TOC offers a lot of promising synergy to TRIZ. Both share the assumption that most core problems exist because some underlying conflict or contradiction prevents straightforward solution of a problem. Each approach is capable to analyze the problem situation and to identify the conflict. TRIZ is especially strong in generating innovative concepts to overcome these conflicts. TOC additionally provides tools to answer the question

»How do we implement the change?«.

The Theory of Constraints is a systems approach to continuous improvement that has its basis in the manufacturing environment. TOC was developed by Eliyahu M. Goldratt in 1984 which he presented in his first book ‘The Goal’. Goldratt likens (production) systems as chains, or networks of chains. A process chain in manufacturing can be thought of as a chain of dependent events that are linked together. The activities that go on in one “link” are dependent upon the activities that occur in the preceding “link”. Since »a chain is only as strong as its weakest link«, optimization efforts should focus on “chain strength” by working to strengthen the weakest link – the constraint. TOC is based on several principles that are very important for successful improvement. The crucial principle is »Systems as Chains« that has already been mentioned. Other vital ones are /Lit 6/:

»Cause and Effect«: All systems operate in an environment of cause and effect. One particular event acts as a cause for another event, while the particular cause leads to a specific effect. This relationship between cause and effect can be very complex.

»Undesirable Effects and Core Problems«: The indication of the existence of a problem is brought out by undesirable effects (symptoms). Eliminating undesirable effects gives a false sense of security. The elimination of the core problem, however, not only eliminates the symptoms but prevents them from happening again.

»Solution Deterioration«: The solution to any problem deteriorates with time, because the environment changes. Hence, a process of continuous improvement is required to maintain the same levels of efficiency at all times.

»Ideas are not Solutions«: Mere the idea on how to solve a problem does not result in improvement, rather, the effective implementation of this idea that results in real improvement. However, in many cases ideas fail in their implementation stage.

The TOC thinking process focuses on the answer to three fundamental questions: »What to change?«, »What to change to?« and »How to cause the change?«. This thinking process is supported by specific thinking tools, the five logical trees (see slide 6). The Current Reality Tree (CRT) is designed to analyze the current condition of a system and to gain a better understanding for the problem. It identifies the core problem(s) that lead(s) to observed undesired effects which decrease(s) the performance of the system. Solving these core problems becomes the objective. This often requires the elimination of an underlying conflict, that prevents straightforward solution. The Conflict Resolution Diagram (CRD), also referred to as “Evaporating Cloud”, helps to resolve these conflicts and strives to create a breakthrough solution to the problem that avoids compromise. Being armed with the tools of TRIZ is here particularly useful. Once a proper improvement measure has been found, the Future Reality Tree (FRT) serves to check if it will in fact produce the desired effect without introducing new and unexpected side effec ts. The FRT can also effectively test alternative solutions before allocating expensive resources to them.

Once the realization of an appropriate solution has become the objective the Prerequisite Tree (PT) comes into operation. The PT is designed to find all obstacles and the responses needed to overcome them in realizing the objective. It identifies minimum necessary conditions and requirements without which the objective cannot be achieved. The result of the PT is a sequence of intermediate objectives to be followed to neutralize all obstacles. The TOC thinking process is completed by the Transition Tree (TT). The TT provides a detailed step-bystep instruction for implementing a course of action. It shows all the steps necessary in achieving a specific objective, providing a so called ‘road map’ to the entire implementation effort. The construction and review of the trees is governed by so-called Categories of Legitimate Reservation (CLR). These are eight rules, or tests of logic that are applied for building, scrutinizing and improving the trees. They also serve to communicate effectively disagreements related to the construction of relationships. The CLRs are clarity, entity existence, causality existence, cause sufficiency, additional cause, cause-effect reversal, predicted effect existence and tautology. /Lit 6/

Both approaches, TRIZ and TOC, provide tools for analyzing the initial situation and for identifying core problems. TRIZ is extremely powerful in generating innovative solutions that overcome underlying conflicts/contradictions and bring the system closer to ideality. The solutions found with TRIZ can be successfully scrutinized and implemented by applying the last three tools of the TOC thinking process (FRT, PT and TT). The combined use of classical QMtools, TRIZ and TOC promises a holistic optimization of process chains in manufacturing, without giving in to compromise. However, to achieve this, two steps were essential. On the one hand some of the individual tools had to be redesigned in order to fit them together. On the other hand a procedure was developed that sets the individual tools into context.

DMAIC – Step-by-Step to Success

The system for innovative process chain optimization that was developed in the research project consists of two major elements. A procedure describing the course of an optimization project and an interdisciplinary toolset that supports each stage of the model. The toolset contains TRIZ-, TOC- as well as classical QM-tools (see slide 5). The procedure was developed on the basis of the DMAIC cycle (Define-Measure-Analyze-Improve-Control) that originates from the Six Sigma approach /Lit 7/.

In Define the team has to identify and define the problem, the objective, customer requirements (internal and external) and important boundary conditions. The process chain to be investigated needs to be understood. To support this stage the Define-Checklist was developed, based on the Innovative Situation Questionnaire of TRIZ. The objective of Measure is to determine the current performance of the process and the extent of the problem. The current state of the process is recorded which is essential to rate the achieved success at the end of the optimization project. The Measure phase also prepares the analyze step of the DMAIC cycle by gathering key data that helps to identify the process constraint. Usually a lot of data (test data, MTM, …) is already available in the manufacturing area. The gathered data is scrutinized in the Analyze phase. This stage can be classified into data analysis and process analysis. The common objective is to identify the root causes of the problem and the constraint. As a result of Analyze, optimization priorities are set up. Classical QM-tools (Capability studies, 7 tools, …) may serve for data acquisition and evaluation. The process analysis can be effectively supported by the CRT.

The optimization problem is solved in the Improve phase. The constraint in the process chain is examined carefully in order to find out how to elevate it. Sub-problems, conflicts and/or contradictions are systematically revealed by applying several powerful TRIZ-based tools (e.g. Function-Effect-Modeling). Innovative concepts for solutions are generated, structured and compared. The solutions to be implemented are selected. This can be done by portfolio analysis (e.g. chance/risk; cost/time) or the application of the Future Reality Tree, for instance. The implementation of these solution is methodical safeguarded by the Prerequisite Tree and Transition Tree. The first 4 steps of DMAIC were dedicated to identifying, measuring and implementing change. However, without sustaining the gain, the initial enthusiasm for improvement can easily be lost. The Control phase serves for confirming the fact that the improvement measures selected have achieved the goal set up in Define. Therefore, compiled result data has to be reviewed. A second important objective of Control is to select ongoing measures to monitor performance of the process and continued effectiveness of the implemented solutions.

Conclusion
Classical methods of quality management in manufacturing already do a good job in the prevention of failures and can optimize the output of an individual process. However, improvement efforts normally focus on individual elements of a process chain and do not question the existing structures. When it comes to improving a process chain with a holistic view to increasing efficiency and effectiveness, the conventional QM-methods often reach their limits. Especially, when the complexity of the system is high, support is needed.

Core problems that prevent straightforward solution are often caused by an underlying conflict or contradiction. For example, from the current point of view, improving one characteristic of the process will result in impairing another characteristic. Hence, a trade-off seems to be necessary. TRIZ provides powerful tools for overcoming conflicts and contradictions without the need for compromise. TRIZ expands the knowledge horizon by providing a knowledge basis that represents the combined experience of over 2.5 million patents. It also helps users to detach themselves from their usual thought patterns and structures.

Armed with TRIZ, the optimization team can generate innovative concepts for breakthrough solutions. However, ideas or concepts are not solutions. Not until they have been successfully implemented. The implementation stage needs to be methodically safeguarded as well. This is where TRIZ and TOC can complement one another. TOC provides tools that enable the user to evaluate alternative concepts for solutions and ensure successful realization. TOC also supports the analysis of the process chain in order to identify the “weakest link”, the constraint, that needs to be strengthened.

The DMAIC cycle as a proven procedure for process optimization sets the individual tools into context. It emphasizes the measurement of the current performance of the process chain, which is essential to evaluate the achieved improvement. The concerted application of classical QM-tools, TRIZ and TOC in the DMAIC cycle results in a holistic optimization of process chains with innovative design. Even if the maturity of the system is high, improvements in quantum leaps are possible. The system for innovative process chain optimization enables enterprises to tap substantial innovation-based optimization potentials and contributes to improving both their quality standards and their competitiveness.