Systematic Clustering of Business Problems

Related Tools & Articles
  • Discussion Forum
    "TRIZ is clumsy for business strategy and dealing with customers' desires."

    Contribute to this Discussion

    By Junyoung Kim and Yongtae Park

    Abstract

    In this research, the traditional engineering area of the Theory of Inventive Problem Solving (TRIZ) is extended to the business area to solve the business contradictions that shape technology evolution – clarifying that TRIZ is valid in the business arena as well as the technical arena, although there is less data. This paper proposes using the 40 inventive principles of TRIZ to solve problems in business environments, systematically and scientifically reproducible. For technology forecasting purposes, it is required to better understand the business contradictions which direct the evolution of technology. The authors compiled 540 business cases (instead of the conventional patents), conducted cluster analysis and developed a TRIZ contradiction table that might be used to solve business problems in the same way as the TRIZ contradiction matrix is used in engineering environments. It is likely that as more cases are analyzed, the more robust the TRIZ contradiction table in business environments will be. Technical and physical contradiction approaches in business are also suggested for use. This paper introduces a clustering method using the TRIZ principles to forecast the technology evolution which is shaped by business. The first step is to identify problem-inducing prameters. The second step is to determine a particular approach – whether the parameters are solely included or used in multiples. The third and final step is to find an idea to solve the problem using the reconstructed TRIZ contradiction table. This paper facilitates this approach with case implications, blue ocean and red ocean strategy, and agile supply chain management cases.

    Keywords

    Clustering, extended TRIZ contradiction table, business case, technical contradiction, physical contradiction

    Introduction

    The TRIZ methodology has been used traditionally to solve engineering problems. Research related to TRIZ explored how it can can be a useful tool for solving contradictions in engineering problems and which principles can be applied to a problem by analyzing patents that are treated as the improved output. The father of TRIZ, Genrich Altshuller, and his colleagues analyzed more than 40,000 patents and identified common patterns to solve contradictions of engineering problems. TRIZ includes a set of he common patterns based on the IFR (ideal final result). The quantity of research studying the applicability of TRIZ to other fields has recently increased. TRIZ tools are integrated and connected with other managerial and engineering tools.29 The most frequently used TRIZ tools are the most basic tools, like problem solving based on the contradiction matrix and predicting based on the evolution of technology.21

    TRIZ is a collection of creativity tools, which enable quantum leaps in technology-focused contradiction problem solving. In addition to the usefulness of TRIZ as a problem solving tool, it is able to forecast technology.27 From the latter's perspective, it is possible to envision technological achievement through the use of the parameters and principles of TRIZ – by solving one problem with TRIZ, the next stage of evolution can be reached. Arguably, it is necessary to understand the business evolution surrounding a technology in order to define the future of the technology.3 The reason is that a business has a higher-level hierachy to the technology itself – the firm and its industry are indicators that guide and direct technology evolution. Technology evolution depends on the capability of firms developing the technology, the competitiveness of the operating firms and the demands of the customers of the firms. Prior to analyzing a particular technology, it is necessary to analyze the business environment; the result of these analyses affect technology evolution prediction.

    The TRIZ methodology guides the problem solver to solve the problem with prior solved sets, represented in highly abstracted forms. There are examples of business case anaylsis prior to this paper:

    In the non-engineering context, by solving contradictions in the business fields, especially in the strategic management context, many managers are trying to derive more competitive advantages than competitors in the same business environment. Managers find the solution to the strategic management problem (by deriving more competitive advantage) within specific constraints (in the same business environment).

    Harvard professor and leading expert on competitive strategy Michael E. Porter has suggested analyzing the strategic attributes of a firm in two ways: value chain analysis and value net analysis, respectively inside and outside of the firm. In value chain analysis, the firm is a set of subsequent processes adding value; in value net analysis, the firm is regarded as an element of the industry in the domain view. Based on previous research, it is possible to step forward to analyze the parameters affecting the problem solving process.8,4 The previous research suggested using the paramters related to the problem in the business context and the TRIZ inventive principles in some contradictable business cases. The authors analyzed each business case and distinguished parameters from the cases and bridged parameters to the TRIZ inventive principles.

    Altshuller defined two types of contradictions – technical and physical. The technical contradiction is easily observed and can be solved by using the 40 inventive principles in the contradiction matrix. In TRIZ, there are also 39 parameters listed in the contradiction matrix. From the matrix, it is necessary to identify two parameters: 1) what needs to be improved and 2) what should not be worsened. In the next step, the two parameters identified previously are used to look up the most common inventive principles that have been adopted by patentees to resolve this contradiction. The contradiction matrix is a 39 x 39 matrix, a subset of which is shown in Figure 1.

     Figure 1: Section of Contradiction Matrix

    Each cell contains some selected principles that guide the problem solver to the proper solution. Whoever can recognize and analyze his problem can solve it, having been shown the way a technical contradiction is solved. But the technical contradiction solving method has some drawbacks. Generally, technical contradictions always have two parameters. As shown in Figure 2, the improvement of parameter A is concatenated to the worsening of parameter B – the easy way to solve these problems is to find the optimal point between them. This way might be a proper measure, but it is not an ideal solution, just a fixation. The physical contradictions are the characteristics related to the origin of a problem. As shown in Figure 3, parameter C is the origin. If the problem solver can find the more fundamental parameter C, parameters A and B can both be improved. This is the physical contradiction solving method. Altshuller identified four directional separation principles, which are used to solve the physical contradiction: space, time, condition and transition. The physical contradiction solving method uses just one parameter so that it covers a more fundamental area than the technical one can. Both methods can be used to analyze common problems that occur in a business context and finally forecast business evolution in the value chain.

     Figure 2: Technical Contradiction


     Figure 3: Physical Contradiction

    While much research has been conducted, empirical evidence is still lacking in the business context. Recent research questions whether TRIZ inventive principles are useful or not.15 This research will help TRIZ tools move forward when the research questions are about the usefulness of the tool ex-ante rather than the ex-post explanation. The other part requiring more research attention is the organizing process of the TRIZ inventive principles. There is no doubt that Altshuller and his colleagues analyzed and built the TRIZ framework, but the processes to analyze patents and build the TRIZ inventive principles are barely known according to some researchers, who insist that TRIZ is not a scientific approach, claim.

    The reproducibility problem also needs further exploration. In this paper the systematic and scienctific reproducible approach is used to solve the business problem by using empirically-conducted business cases as data. The approach can be replicated – and if it is replicated with more data, more robust results are predicted.

    Research Design

    The four steps of the full process of research is shown in Figure 4.

     Figure 4: Research Flow

    Step 1: Vector representation of 540 business cases

    Enginneering problems and business problems are different in hierarchy. For example, business problems are often more abstract than engineering problems. The authors employed the 31 parameters in the business environment developed by Darrell Mann.5 The business problem and the solution are expressed by the business case. Research has matched a number of business cases and the inventive principles used with them.7,18 The authors analyzed 540 business cases excluding ambiguous sayings and proverbs. They were compiled by management best practices into 32 dimension vectors. The thirty-one dimensions are the parameters related to the business cases, while the last is one of the TRIZ inventive principles that would be a useful guide to solve the problem, the business case.

    Step 2: Clustering business cases with TRIZ principles and dominant parameters

    The authors adopted the systematic and scientific approach in utilizing the concept of the TRIZ inventive principles. The 31x31 TRIZ contradiction matrix in business context has already been constructed and is available in a software suite – the CREAX Innovation Suite. The construction processes, thus, are not available to amateurs – a rebuilt table specific to business problems that is reproducible is scientifically valuable. For this purpose, clustering analysis was conducted (according to the problem-inducing parameters) with the 540 business problems.

    The clustering analysis is conducted with two steps of hierarchical clustering analysis: 1) hierarchical clustering analysis method Ward's linkage and 2) a non-hierarchical step called k-means clustering. Among the hierarchical analyses, Ward's linkage function specifies that the distance between two clusters is computed as the increase in the error sum of squares (ESS) after merging two clusters into a single cluster. Ward's method seeks to choose the successive clustering steps so as to minimize the increase in ESS at each step; this method can be used to discover the proper number of clusters. K-means clustering analysis is employed to allocate the 540 business cases into clusters. K-means clustering is one of the simplest unsupervised learning algorithms, which assigns the best data sets using known numbers of clusters. The parameter or parameters that are related to an entire business case within a cluster is here called the "dominant parameter" and the clusters are reorganized according to the dominant parameters as "strategic clusters." Some clusters have no dominant parameter; it was concluded that this resulted due to data shortage.

    Step 3: Building a TRIZ principles table in the manner of technical contradictions

    The dominant parameter per cluster is regarded as the parameter that mainly invokes the problem. In the CREAX Innovation Suite, the comparison used here, it is suggested that TRIZ principles be related to only two parameters like conventional engineering's TRIZ contradiction matrix. This research found that some business case clusters involve four parmeters (the maximum) and some clusters involve only one. The first strategic cluster involving four parameters about an interface problem was compared to CREAX's result. According to the result of the Wilcoxon signed rank test, it was concluded that the difference between the two results is statistically insignificant. In this manner, the TRIZ inventive principles in business context are similar to the traditional and that of the CREAX Innovation Suite.

    Step 4: Building an extended TRIZ principles table in the manner of physical contradictions

    It has been suggested that the table linking the TRIZ inventive principles to the four principles could be used to solve the physical contradiction in a business context.5 By integrating the previous results, the physical contradiction problem solving method was added to the technical contradiction table only.

    Data and Analysis

    Data

    Thirty-one parameters were observed by Darrell Mann as common patterns in solving business problems.5 (See Table 1) They are based on the work of quality guru W. Edwards Deming, which distinguished processes that are sequential and parts of the production are a product. The processes are the initial research and development and pre-production activities, the production process, the supply process and post-supply support activities. These processes can be considered as the physical parameters of time, cost and risk which are element parameters in the business context.5 In addition, there exists an interface element that interconnects other elements. Based on these elements, Mann developed 20 parameters from four production processes (R&D, production, supply, support) and each has five parameters. In the business literature and the problem solving process, 11 supportive parameters were added.5 These 31 parameters are used in the CREAX Innovation Suite.

    Table 1: 31 Parameters Frequently Observed in Business Problems

    Number

    Parameter

    1

    R&D spec/capability/means

    2

    R&D cost

    3

    R&D time

    4

    R&D risk

    5

    R&D interfaces

    6

    Production

    7

    Production cost

    8

    Production time

    9

    Production risk

    10

    Production interfaces

    11

    Supply

    12

    Supply cost

    13

    Supply time

    14

    Supply risk

    15

    Supply interfaces

    16

    Product reliability

    17

    Support cost

    18

    Support time

    19

    Support risk

    20

    Support interfaces

    21

    Customer revenue/demand/feedback

    22

    Amount of information

    23

    Communication flow

    24

    System-affected harmful effects

    25

    System-generated side effects

    26

    Convenience

    27

    Adaptability/versatility

    28

    System complexity

    29

    Control complexity

    30

    Tension/stress

    31

    Stability

    In recent research, the availability of the TRIZ inventive principles in fields exclusive of engineering area has been analyzed, and those cases have been used as a research basis here.7,18 Each case is represented as a linear combination of the 31 parameters developed by Mann, as shown in Figure 5. If the parameter is used to solve the problem, then the vector value of the parameter dimension is assigned one (1); if the parameter is not used in the case then the vector value is assigned zero (0). The criteria used are based on business administration literature.14,22,23,24,1,9,25,17 In this way, all cases are represented with Mann's originally suggested 31-dimensional vectors. The last column indicates the TRIZ inventive principle used to solve the case. For example, the 32nd column has a value of a range from 1 to 40 according to the case-related principle.7,18 (Sayings and proverbs are suggested in the earlier research, but for this article's purposes the parameters seemed ambiguous and easily misunderstood, so they are excluded here.)

    In practice, real business problems have more than one contradiction between two parameters. For finding a solution, the best practice consists of many parameters connected to the problem. For instance, one article suggests a TRIZ solution case in which the best practice, the Lean production method, is related to four parameters: R&D time, R&D cost, production quality and production cost.2 The solution is the optimal one that sets up the trade-offs among the four parameters. In separate research, the Lean production method is named called Lean manufacture, and is shown to use the second of the 40 principles, taking out.7 The Lean production method is shown in relation to the 32-dimension vectors in Figure 5.

     Figure 5: Vector Representation of Lean Production Method

    Clustering Analysis

    Before categorizing the cases into strategic clusters, the authors performed hierarchical clustering analysis to figure out the proper number of clusters using Ward's linkage method. The result of the analysis is shown in Figure 6. The result leads to the conclusion that the 20 clusters are the best because of the similarity (or correlation coefficient) larger than 50. To allocate the cases into the clusters, the k-means clustering method was used. This method allocated each case into the nearest cluster, which compared the distances (dissimilarity of cases) of the neighboring centers of the strategic clusters. The result of the k-means clustering method is shown in Table 2. The minimum number of cases contained in the strategic cluster is two, and the maximum has ninety-two cases.

     Figure 6: Ward's Linkage

     

    Table 2: K-means Clustering Analysis Results

    Cluster

    Number of Observations

    Within Cluster Sum of Squares

    Average Distance from Centroid

    Maximum Distance from Centroid

    1

    24

    19.750

    0.859

    1.735

    2

    92

    60.870

    0.743

    1.456

    3

    8

    3.875

    0.619

    1.068

    4

    2

    0.500

    0.500

    0.500

    5

    2

    0.500

    0.500

    0.500

    6

    5

    3.200

    0.785

    0.980

    7

    6

    2.333

    0.601

    0.972

    8

    4

    2.000

    0.707

    0.707

    9

    4

    1.750

    0.645

    0.901

    10

    14

    15.071

    1.020

    1.361

    11

    71

    69.408

    0.973

    1.417

    12

    50

    39.280

    0.803

    1.528

    13

    7

    4.857

    0.783

    1.262

    14

    29

    18.552

    0.769

    1.293

    15

    3

    3.333

    1.033

    1.247

    16

    4

    1.500

    0.530

    1.061

    17

    91

    66.703

    0.781

    1.976

    18

    62

    12.468

    0.267

    1.394

    19

    8

    2.375

    0.479

    0.910

    20

    54

    2.907

    0.093

    0.982

    If all allocated cases in a strategic cluster are related to a specific parameter, it is a dominant parameter; if only some of the cases are related to a specific parameter, it is a partial parameter. A dominant parameter means that a specific parameter should be included in solving certain cases in a strategic cluster. As shown in Figure 7, cluster 1 has 24 business cases, which are related to the parameters 5, 10, 15 and 20, and are dominant parameters. From the perspective of k-means clustering, the dominant parameter is the center of the cluster and the place where the common contradiction happens. But parameter 21 is only related to cases 15 and 24 – this is a partial parameter. All of the dominant parameters (5, 10, 15 and 20) are related to the interface; this cluster is the principle to solve the technical contradiction that occurs in the field of the interface between business activities. This process is repeated for each cluster and shown in Table 3. The 20 clusters are connected to the dominant parameters as business problem-solving principles that are also cluster centers.

     Figure 7: Strategic Cluster 1

    The dominant parameters of cluster 1 are R&D interfaces, production interfaces, supply interfaces and support interfaces. All of the dominant parameters are related to the interface problem. The number of cases allocated into cluster 1 is 24. The TRIZ inventive principles necessary to solve these cases are 2, 5, 6, 15, 26, 28, 29, 33, 38 and 40. Clusters related to none of the dominant parameters are 6, 9, 11, 24, 25 and 26. These sparse clusters may be caused by a shortage of cases being studied are not enough to treat as meaningful data.

    Table 3: 20 Strategic Clusters with Dominant Parameters

    Cluster

    Dominant Parameters

    Number of Parameters

    Number of Cases

    TRIZ Inventive Principles

    1

    R&D interfaces, production interfaces

    4

    24

    2, 5, 6, 15, 26, 28, 29, 33, 38, 40

    2

    Customer

    1

    92

    1, 2, 3, 4, 5, 6, 7, 13, 14, 15, 16, 17, 19, 22, 24, 25, 27, 29, 30, 31, 35, 36, 38, 40

    3

    Production interfaces, support

    2

    8

    1, 3, 6, 33, 40

    4

    R&D interfaces, customer

    2

    2

    1, 4

    5

    Support interfaces, customer

    2

    2

    1, 16

    6

    None

    0

    5

    2, 10, 18, 34

    7

    R&D risk

    1

    6

    11, 22, 23, 35

    8

    R&D interfaces, support

    2

    4

    1, 6, 40

    9

    None

    0

    4

    1, 2, 23, 38

    10

    R&D spec/capability/means

    1

    14

    1, 4, 11, 13, 14, 17, 19, 23, 25

    11

    None

    0

    71

    1, 2, 4, 5, 6, 7, 8, 9, 19, 13, 14, 15, 16, 17, 18, 20, 22, 23, 25, 28, 30, 31, 32, 35, 40

    12

    Production risk

    1

    50

    1, 5, 10, 11, 13, 17, 20, 22, 23, 25, 26, 33, 40

    13

    Product reliability

    1

    7

    1, 11, 19

    14

    Support risk

    1

    29

    1, 11, 13, 14, 15, 25, 27, 28, 30, 33, 35, 40

    15

    Supply risk, supply interfaces

    2

    3

    5

    16

    Production cost, production time

    4

    4

    1, 15, 24, 27

    17

    Tension/stress

    1

    91

    1, 2, 3, 8, 9, 11, 17, 24, 28, 35, 39

    18

    Support interfaces

    1

    62

    1, 3, 5, 6, 7, 10, 17, 23, 24, 25, 28, 30, 33, 38, 40

    19

    Stability

    1

    8

    3, 29, 36

     20

    None

    0

    54

    1, 2, 4, 5, 6, 7, 9, 10, 12, 13, 14, 15, 17, 18, 19, 20, 21, 24, 26, 27, 32, 33, 34, 37, 39

    Building a TRIZ Contradiction Table

    Technical contradictions are involved in more than two parameters in this business cases, differing in parameter numbers from the CREAX Innovation Suite. To evaluate the validity of the strategic cluster, it was compared to the guided problem solving principles of the CREAX Innovation Suite. The CREAX-generated principles for solving the interface problem are shown in Table 4. Except for the frequency of occurrence, the results are quite similar. The CREAX Innovation Suite generated the principles 2, 5, 6, 15, 26, 28, 29, 33, 38 and 40, which are also proposed by the authors' work. The constitution processes of the CREAX Innovation Suite are not known, but the result is similar to the TRIZ contradiction matrix in the traditional engineering context. Each cell in the 31x31 matrix has the TRIZ principles which are used to solve the problem suggested by the crossed parameters. As shown in Table 4, the TRIZ inventive principles 28, 40, 6 and 29 might be used to solve the problem with the trade-offs among R&D interfaces and production interfaces. According to the statistical analyses conducted, all interface parameters are involved in the business problem, which might be solved with the noted ten inventive principles.

    Table 4: Comparison Between CREAX's and the Newly Proposed TRIZ Principles for Solving Interface Problems

    Tools

    Parameters

    Inventive Principles

     

    All used

    2, 5, 6, 15, 26, 28, 29, 33, 38, 40

    CREAX Innovation Suite

    R&D interface/production interface


    R&D interface/supply interface


    R&D interface/support interface


    Production interface/supply interface


    Production interface/support interface


    Supply interface/support interface

    28, 40, 6, 29


    28, 40, 6, 15, 29


    28, 40, 6


    38, 5, 2, 26


    40, 33, 6


    5, 6, 38, 40

    Proposed

    Strategic cluster 1:
    R&D interface, production interface,
    supply interface, support interface

    2, 5, 6, 15, 26, 28, 29, 33, 38, 40

    These two approaches are different methodologically-speaking, which was validated from this result. For accurate validation, the Wilcoxon signed-rank test (comparisons between different measurements) comparing the occurrence frequencies of the two approaches was run. Each pair is about the principle used in the approach. The p-value is 0.933 and it is concluded that the difference between the two approaches is not statistically significant. It is also concluded that it is a better choice to adopt the authors' approach to solve the technical contradiction than that of the CREAX system, because of its easy reproducibility.

    The results have been reorganized as shown in Table 5, to show the TRIZ inventive principles that might be used to solve the problems specified by the problem-inducing parameters. As the traditional TRIZ contradiction matrix is used to solve engineering problems, Table 5 can be used to solve business problems. The first step is to identify the confronting business problem. The second step is to analyze the business problem and find the problem-inducing parameters out of the 31 parameters. The third step requires the use of the information shown in Table 5, finding the TRIZ inventive principles that are related to the parameters and solving the business problem.

    Table 5 has a limitation. Some of the parameters do not have corresponding principles; there is a question about whether the allocated TRIZ inventive principles are proper. The authors again attribute the cause to the shortage of data. More research is likely to be conducted in this manner to make Table 5 more robust.

    Table 5: TRIZ Contradiction Table in Business Context

    Number

    Parameter

    Other Parameters

    Useful Inventive Principles

    1

    R&D spec/capability/means

    None

    1, 4, 11, 13, 14, 17, 19, 23, 25

    2

    R&D cost

    N/A

     

    3

    R&D time

    N/A

     

    4

    R&D risk

    None

    11, 22, 23, 35

    5

    R&D interfaces

    10, 15, 20


    20


    21

    2, 5, 6, 15, 26, 28, 29, 33, 38, 40


    1, 6, 40


    1, 4

    6

    Production

    N/A

     

    7

    Production cost

    8, 9, 10

    1, 15, 24, 27

    8

    Production time

    7, 9, 10

    See number 7, production cost

    9

    Production risk

    7, 8, 9

    See number 7, production cost

    10

    Production interfaces

    5, 15, 20


    7, 8, 9


    20

    See number 5, R&D interfaces


    See number 7, production cost


    1, 3, 6, 33, 40

    11

    Supply spec/capability/means

    N/A

     

    12

    Supply cost

    N/A

     

    13

    Supply time

    N/A

     

    14

    Supply risk

    15

    5

    15

    Supply interfaces

    5, 10, 20


    14

    See number 5, R&D interfaces


    See number 14, supply risk

    16

    Product reliability

    None

    1, 11, 19

    17

    Support cost

    N/A

     

    18

    Support time

    N/A

     

    19

    Support risk

    None

    1, 11, 13, 14, 15, 25, 27, 28, 30, 33, 35, 40

    20

    Support interfaces

    None


    5, 10, 15


    10


    21

    1, 3, 5, 6, 7, 10, 17, 23, 24, 25, 28, 30, 33, 38, 40


    See number 5, R&D interfaces


    See number 10, production interfaces


    1, 16

    21

    Customer revenue/demand/feedback

    None


    20

    1, 2, 3, 4, 5, 6, 7, 13, 14, 15, 16, 17, 19, 22, 24, 25, 27, 29, 30, 31, 35, 36, 38, 40


    See number 20, support interfaces

    22

    Amount of information

    N/A

     

    23

    Communication flow

    N/A

     

    24

    System-affected harmful effects

    N/A

     

    25

    System-generated side effects

    N/A

     

    26

    Convenience

    N/A

     

    27

    Adaptability/versatility

    N/A

     

    28

    System complexity

    N/A

     

    29

    Control complexity

    N/A

     

    30

    Tension/stress

    None

    1, 2, 3, 8, 9, 11, 17, 24, 28, 35, 39

    31

    Stability

    None

    3, 29, 36


    Building an Extended TRIZ Contradiction Table

    The case of physical contradiction solutions is more complicated, because only one parameter is related to the problem. The authors analyzed the spatial categorical correlation between principles used to solve the physical contradiction suggested in Mann's 2002 research, and the selected clusters among the 20 strategic clusters – that involve more than 10 cases and at least one dominant parameter.5 These are seven clusters – 1, 2, 10, 12, 14, 17 and 18. All of these, except cluster 1, involve only one dominant parameter. The relationships among the physical contradiction in the business context and the TRIZ inventive principles are shown in Table 6. Mann proposed that when the business problems can be located and mapped in the real territory, this resolution might induce the contradiction to emerge as a physical contradiction.5 The similarity of the TRIZ principles to solve the physical contradiction and the TRIZ principles used in the result of the previous work was considered. The Ward linkage and correlation coefficient distance is used to test their similarity.

    Table 6: Proposed Physical Contradictions in Business Context5
    SeparationMatched TRIZ Principles
    Space1, 2, 3, 4, 7, 13, 14, 24, 26, 30, 37
    Time1, 9, 10, 11, 15, 16, 18, 19, 20, 21, 29, 34
    Condition12, 28, 31, 32, 35, 36, 38, 39, 40
    Transition1, 5, 6, 7, 8, 13, 22, 23, 25, 27, 35

    Physical contradictory parameters and the clusters are matched in the same way as hierarchical clustering criteria. A dendrogram produced with Ward linkage and correlation coefficient distance calculation is used. The related clusters and the TRIZ inventive principles are shown in Table 7. If confronted by the problem whose characteristics are related to cluster 2, the problem can be solved by using the principle of "separation in space," which is improved over the trade-off result. In the same way, cluster 17 is resolved with the principle of "separation in condition" and clusters 10, 12 and 14 are related to the principle of "separation in transition." No clusters that could be solved with "separation in time" were found; this may also be due to a shortage of studied cases. The clusters and principles were mapped into three-dimensional space by using MDS (multi-dimensional scaling). This is shown in Figure 8. MDS can be considered an alternative to factor analysis. Multi-dimensional visualization helps to see which clusters are related to a particular principle.

     Figure 8: MDS Result

    Table 7: Proposed Physical Contradictions in Business Context

    Separation

    Related Clusters

    Matched TRIZ Principles

    Space

    Cluster 2

    1, 2, 3, 4, 5, 6, 7, 13, 14, 15, 16, 17, 19, 22, 24, 25, 27, 29, 30, 31, 35, 36, 38, 40

    Time

    N/A

    N/A

    Condition

    Cluster 17

    1, 2, 3, 8, 9, 11, 17, 24, 28, 35, 39

    Transition

    Clusters 10, 12, 14

    1, 4, 5, 10, 11, 13, 14, 15, 17, 19, 20, 22, 23, 25, 26, 27, 28, 33, 35, 40

    The authors extended the results of Table 5 based on the results of Table 7. In Table 5, it is fair to say that the useful TRIZ inventive principles might be used to solve the business problem, and that these problems usually contain more than two parameters. The used TRIZ principles were added to Table 5 when the number of parameters involved is only one – like the physical contradiction in the traditional engineering context. As shown in Table 7, when the problem is indentified to be in strategic cluster 2, the separation in space principle can be used. A dominant parameter of strategic cluster 2 is "customer revenue/demand/feedback" – when the origin of the problem is the parameter, the separation in space can be used. The "tension/stress" parameter of strategic cluster 17 is related to the "separation in condition" principle for solving the physical contradiction; the "R&D spec/capability/means" parameter is related to strategic clusters 10, 12 and 14 with "separation in transition" for solving the physical contradiction. These results are supplemental to Table 5 and represented in Table 8.

    Table 8: Extended TRIZ Contradiction Matrix in Business Context

    Number

    Parameter

    Another Parameter

    Useful TRIZ Principles

    1

    R&D spec/capability/means

    None

    Separation in transition

    2

    R&D cost

    N/A

     

    3

    R&D time

    N/A

     

    4

    R&D risk

    None

    11, 22, 23, 35

    5

    R&D interfaces

    10, 15, 20


    20


    21

    2, 5, 6, 15, 26, 28, 29, 33, 38, 40


    1, 6, 40


    1, 4

    6

    Production spec/capability/means

    N/A

     

    7

    Production cost

    8, 9, 10

    1, 15, 24, 27

    8

    Production time

    7, 9, 10

    See number 7, production cost

    9

    Production risk

    None


    7, 8, 9

    Separation in transition


    See number 7, production cost

    10

    Production interfaces

    5, 15, 20


    7, 8, 9


    20

    See number 5, R&D interfaces


    See number 7, production cost


    1, 3, 6, 33, 40

    11

    Supply spec/capability/means

    N/A

     

    12

    Supply cost

    N/A

     

    13

    Supply time

    N/A

     

    14

    Supply risk

    15

    5

    15

    Supply interfaces

    5, 10, 20


    14

    See number 5, R&D interfaces


    See number 14, supply risk

    16

    Product reliability

    None

    1, 11, 19

    17

    Support cost

    N/A

     

    18

    Support time

    N/A

     

    19

    Support risk

    None

    Separation in transition

    20

    Support interfaces

    None


    5, 10, 15


    10


    21

    1, 3, 5, 6, 7, 10, 17, 23, 24, 25, 28, 30, 33, 38, 40


    See number 5, R&D interfaces


    See number 10, production interfaces


    1, 16

    21

    Customer revenue/demand/feedback

    None


    20

    Separation in space


    See number 20, support interfaces

    22

    Amount of information

    N/A

     

    23

    Communication flow

    N/A

     

    24

    System-affected harmful effects

    N/A

     

    25

    System-generated side effects

    N/A

     

    26

    Convenience

    N/A

     

    27

    Adaptability/versatility

    N/A

     

    28

    System complexity

    N/A

     

    29

    Control complexity

    N/A

     

    30

    Tension/stress

    None

    Separation in condition

    31

    Stability

    None

    3, 29, 36

    The approach employed to solve the business problem in the manner of physical contradictions is simpler than the engineering approach, but the TRIZ approach has been modified into a method used to solve business problems.

    Case Implications

    Mann's research expanded upon Deming's work to enlarge the applicable area of TRIZ from an engineering context to a business context.5 Deming's view on firms is quality management-centered. In quality management thinking, a firm is an organization of processes that can be improved through finding the most efficient and optimal operation. But quality management is a part of overall business management and has limitations when it is necessary to solve other problems. Mann added parameters 21 to 31 cover all possible problems. As a result, this earlier application of TRIZ to business problems led to an organized, overall view of firm activities, processes and operation domains similar to the value chain.5

    The authors then mapped the derived strategic clusters on the value chain. In the value chain perspective, the technical contradiction can be solved to set the optimal point of the trade-offs that happen between steps of the problem. The derived strategic clusters are located in each process, and between processes, of the value chain. As Figure 9 shows, for instance, cluster 2 includes aggregated TRIZ inventive principles that are used to solve the problem related to the "customer revenue/demand/feedback" parameter and can be placed in the process "marketing/sales." Only the clusters with more than 10 cases and at least one dominant parameter are mapped, based on the authors' collected and analyzed work.

    The business context derives some of the physical contradiction solving principles: space, condition and transition. These physical contradictions are inherent to the collected cases pool.

     Figure 9: Value Chain with Clusters of Problem-inducing Parameters and Useful TRIZ Principles

    The perspectives of both technical and physical contradictions are adopted to solve whole problems with the TRIZ approach. The problem is classified into four types (space, time, condition and transition) and the classified problem can be solved with the same solution derived from the four types. The same approach was suggested in that the blue ocean strategy is organized with TRIZ inventive principles 1, 8 and 35, and that these can be used to solve the trade-off problem between the "customer revenue/demand/feedback" dominant parameter and other partial parameter.28 This is one of the problems classified into cluster 2 in this research; cluster 2 contains the TRIZ principles 1 and 35, so that the blue ocean strategy work validates this research. By using blue ocean strategy, finding new profitable markets can be achieved through conducting the separation analysis between red and blue oceans at an abstract level. The separation analysis is used with the separation principles used to solve the physical contradiction.

    There is an agile supply chain management case that relates to the case using the physical contradiction method.19 It was concluded that companies confronting demand uncertainty adopted agile supply chain management. In that research, the firms producing highly innovative products suffer unpredictable demand caused by short lifecycle products with new technology and trend content. These firms have to equip themselves with the ability to respond to demand immediately and produce with agility. This kind of business problem is involved in the problem inducing parameter 21, "customer revenue/demand/feedback;" parameter 21 is the only parameter involved. As proposed earlier in this paper, with the help of Table 8, the separation in space principle might be used to solve this kind of problem. The agile supply chain management case analyzed that high-tech firms confronting demand uncertainty such as Xilinx and Adaptec used agile supply chain management.19

    "Xilinx Inc., a fab-less semiconductor company specializing in high-end integrated circuits known as field-programmable logic … Since they are pushing the frontier, the process technology used and process control methods required for the wafer fabrication process are very demanding and challenging. A highly sophisticated fabrication facility is needed. Xilinx has formed very tight partnerships with two such foundries, United Microelectronics Corporation in Taiwan and Seiko in Japan. Fabricated wafers are then stocked, forming a decoupling point known as die banks. As demand for specific chips is known through orders from customers such as Cisco, Dell, Motorola, HP, and Lucent Technologies, the final assembly and testing of the chips are carried out by other supply chain partners in Korea and the Philippines. Such a decoupling point strategy enables Xilinx to be responsive to the diverse and changing needs of their customers, who themselves are faced with highly unpredictable demand for their end products.

    "Adaptec, another fab-less semiconductor company faced with both evolving supply processes and innovative products, also relied on advanced Internet-based solutions to exchange information and coordinate their production plans with their supply chain partners. Using software… the company communicates in real time with their foundry (Taiwan Semiconductor Manufacturing Company) and their assembly partners (Amkor, ASAT, and Seiko) with information such as detailed and complex design drawings, prototype plans, test results, and production and shipment schedules. This greatly facilitates their ability to be aware of demand and supply levels, and they can respond quickly to potential mismatch problems…"19

    Xilinx used a "decoupling point" supply chain method that can be seen in Wal-Mart and Dell computer cases – the more common term for this is a hub-and-spoke supply chain. These firms confronted demand uncertainty and located the supply hub near the customer to solve their problem. Placing supplies near the customer makes the component, goods and products supply more stable and reliable – even in an unpredictable demand situation. The method adopted by the Adaptec is also frequently used – using software or the Internet. Through the use of software and the Internet, firms are more likely to respond faster. Cross docking systems, barcode systems, EDI (electronic data interface) systems and POS (point of sales) systems are all innovative management tools of Wal-Mart. These firms confronting demand uncertainty are common in that they remove uncertainty using the separation in space principle. Xilinx separates the physical space between some part of the supply chain and Adaptec adopted separated space by using a virtual space – the Internet.

    Discussion

    The research conducted for this paper concludes that the traditional engineering area of TRIZ can be expanded to the business area to solve business contradictions that direct technology evolution. The method is statistically-based and scientifically reproducible. The authors expects that as more business cases are analyzed a more complete and robust result will be achived by evaluating the larger base of empirical data. This paper also suggests that a different approach to solve two TRIZ contradictions is needed to solve businss problems. In the business context, both technical and physical contradiction-solving principles need to be utilized.

    For illustrative purposes, the authors mapped the problem and solution pairs and clusters onto the value chain. By identifying which activity is related to the problem, and how many parameters are related, it is possible to generate ideas to solve a business problem through the problem and solution pairs and clusters located each activity.

    Based on the analyzed business cases, there are business problems with answers and common patterns found through the application of inventive principles. The authors constructed the strategic clusters using clustering methodology. Altshuller and his colleagues analyzed thousands of patents in a quantitative way. But the authors adopted a different method, in which a number of business cases were analyzed and classified with respect to common problem solving patterns in a qualitative and statistical way. This approach shows more accurate and clear ways to the solution, and will improve as more business cases are analyzed – more business cases will offer more statistically significant results. To maintain such data's integrity, the authors suggests that the input value should entered by experts and the values should be weighted.

    The beneficial features of TRIZ has expanded to the business area over the traditional engineering area. Based on previous innovative research, the authors implemented a more advanced model to aid in solving business problems.5,6 This research can be employed to the ex-post validation process not to the ex-ante case. It is expected that more concrete strategic clusters will be used to solve contradictions in business contexts.

    References

    1. Afuah, Allan, (2002), Innovation Management: Strategies, Implementation and Profits, Oxford University Press, USA.
    2. Campbell, Brian, (2004), "Lean TRIZ," The TRIZ Journal, June 2004.
    3. Clarke, Sr., Dana W.,  (2000), "Strategically Evolving the Future: Directed Evolution and Technological Systems Development," Technological Forecasting and Social Change, Vol. 64, pp. 133-153.
    4. Mann, Darrell (2001), "Laws of System Completeness," The TRIZ Journal, May 2001.
    5. Mann, Darrell (2002), "Systematic Win-win Problem Solving in a Business Environment," The TRIZ Journal, May 2002.
    6. Mann, Darrell (2003), "Better Technology Forecasting Using Systematic Innovation Methods," Technological Forecasting and Social Change, Vol. 70, pp. 779-795.
    7. Mann, Darrell, Ellen Domb (1999a), "40 Inventive (Business) Principles With Examples," The TRIZ Journal, September 1999.
    8. Mann, Darrell, Ellen Domb (1999b), "Business Contradictions – 1) 'Mass Customization'," The TRIZ Journal, December 1999.
    9. Teece, David J. (2002), Managing Intellectual Capital: Organizational, Strategic, and Policy Dimensions, Oxford University Press, USA.
    10. Averboukh, Elena A. (2006), Six Sigma Trends: Six Sigma in Financial Services, The TRIZ Journal, April 2006.
    11. Altshuller, Genrich (1999), Innovation Algorithm: TRIZ, Systematic Innovation and Technical Creativity, Technical Innovation Center, Inc.
    12. Altshuller, Genrich, H. Altov, Lev Shulyak (1996), And Suddenly the Inventor Appeared: TRIZ, the Theory of Inventive Problem Solving, Technical Innovation Center, Inc.
    13. Altshuller, Genrich, Lev Shulyak, Steven Rodman (1998), 40 Principles: TRIZ Keys to Technical Innovation, Technical Innovation Center, Inc.
    14. Dosi, Giovanni, David J. Teece, Josef Chytry (1998), Technology, Organization, and Competitiveness: Perspectives on Industrial and Corporate Change, Oxford University Press, USA.
    15. Cong, He, Loh Han Tong (2008), "Grouping of TRIZ Inventive Principles to Facilitate Automatic Patent Classification," Expert Systems with Applications, Vol. 34, No. 1, pp. 788-795.
    16. Lee, Hong Suk, Kyeong-Won Lee (2003), "Practical Case Study of Resolving the Physical Contradiction in TRIZ: Super WaterSaving Toilet System Using Flexible Tube," The TRIZ Journal, November 2003.
    17. Tidd, Joe, John Bessant, Keith Pavitt (2005), Managing Innovation: Integrating Technology, Market and Organizational Change, Wiley.
    18. Zhang, Jun, Kah-Hin Chai, Kay-Chuan Tan (2003), "40 Inventive Principles with Application in Service Operations Management," The TRIZ Journal, December 2003.
    19. Lee Hau L. (2002), "Aligning Supply Chain Strategies with Product Uncertainties," California Management Review, Vol. 44, No. 3, Spring 2002, pp. 105-119.
    20. Loh, Han Tang, He Cong, Shen Lixiang (2006), "Automatic Classification of Patent Documents for TRIZ Users, World Patent Information, Vol. 28, pp. 6-13.
    21. Moehrle, Martin G. (2005), "How Combinations of TRIZ Tools are Used in Companies – Results of a Cluster Analysis," R&D Management, Vol. 35, No. 3, pp. 285-296.
    22. Porter, Michael E. (1998a), Competitive Strategy: Creating and Sustaining Superior Performance, Free Press.
    23. Porter, Michael E. (1998b), Competitive Strategy: Techniques for Analyzing Industries and Competitors, Free Press.
    24. Porter, Michael E. (2004), Competitive Strategy, Free Press.
    25. Bugelman, Robert A., Clayton M. Christensen, Steve C. Wheelwright, Modesto A. Maidique (2003), Strategic Management of Technology and Innovation, McGraw Hill Higher Education.
    26. Mueller, Sandra (2005), "The TRIZ Resource Analysis Tool for Solving Management Tasks: Previous Classifications and their Modification," Creativity and Innovation Management, Vol. 14, No. 1, March 2005.
    27. Technology Futures Analysis Methods Working Group (2004), "Technology Futures Analysis: Toward Integration of the Field and New Methods," Technological Forecasting and Social Change, Vol. 71, pp. 287-303.
    28. Hsiao, Yung-Chin (2005), "Creative Solutions from TRIZ for the Business Contradiction in Red Ocean Strategy," The TRIZ Journal, October 2005.
    29. Hua, Zhongsheng, Jie Yang, Solomani Coulibaly, Bin Zhang (2006), "Integration TRIZ with Problem-solving Tools: A Literature Review from 1995 to 2006," International Journal of Business Innovation and Research, Vol. 1, No. 1, pp.111 -128.

    About the Authors:

    Junyoung Kim received his B.S. degree at Yonsei University with a major in mechanical engineering and and M.S. degree in KAIST (Korea Advanced Institute of Science and Technology) with a major in precision engineering and mechatronics. After graduate school, he worked as an R&D engineer and chief technology officer for 11 years in Samsung Electronics Co. Ltd. He is a Ph.D. candidate in the Interdisciplinary Program of Technology Management at Seoul National University. He is a manager at the Regional Technology Transfer Center at Incheon and is a certified Six Sigma Black Belt at Samsung Electronics Co. Ltd. Contact Junyoung Kim at jykim01 (at) snu.ac.kr.

    Yongtae Park is a faculty member of the Department of Industrial Engineering at the Seoul National University (SNU). He holds a B.S. in industrial engineering from SNU, and an M.S. and Ph.D. in operations management from the University of Wisconsin-Madison. His research interests lie in the areas of industrial knowledge network, knowledge management system, new product/service development and online business modeling. Dr. Park written for numerous publications including Technovation, International Journal of Production Research, Decision Sciences, Technology Analysis and Strategic Management, R&D Management, Technology in Society, Technology Forecasting and Social Change. Contact Yongtae Park at parkyt (at) cybernet.snu.ac.kr.

    Copyright © 2006-2011 – RealInnovation.com, CTQ Media. All Rights Reserved
    Reproduction Without Permission Is Strictly Prohibited – Request Permission


    Publish an Article: Do you have a innovation tip, learning or case study?
    Share it with the largest community of Innovation professionals, and be recognized by your peers.
    It's a great way to promote your expertise and/or build your resume. Read more about submitting an article.