Notes

Outline

Ten Steps to Robust Decisions: Building Consensus and using iDecision in Product Development and Business David G. Ullman Professor, Department of Mechanical Engineering Camas Oregon State University 800 NW Starker Ave Corvallis Oregon, 97331 Corvallis 97330 541-737-2336 541-738-8701 ullman@engr.orst.edu ullman@camas.org www.engr.orst.edu/~ullman

10 Steps to Robust Decisions

Robust decision-making works to minimize the risk of choosing a poor alternative The risk of not being able to solve the problem or     Problem Solver Risk The risk of not getting the best from the problem solvers or Organizational Risk The risk of solving the wrong problem or               Envisioning Risk The risk of not developing good alternatives or          Ideation Risk The risk of choosing a poor alternative or              Evaluation Risk The risk of not following a beneficial strategy or       Strategic Risk The risk of not being able to implement the decision or Execution Risk

Problem Solver Risk The risk of not being able to solve the problem Individual Decision-Making Success = Cognitive Ability + Cognitive Style + Decision-Making Knowledge + Problem Knowledge + Available Resources

Information-processing system

Individual Decision-Making.

Implications of Model Every person's understanding of the problem is different. We recall previous solutions to similar problems or sub-problems to form alternatives. This can get us stuck on a single alternative. The ability to evaluate is dependent on the knowledge about the alternatives.

Cognitive Style PROBLEM SOLVING STYLE INFORMATION MANAGEMENT STYLE DELIBERATION STYLE DECISION CLOSURE STYLE

Slide 9

Slide 10

Slide 11

Slide 12

Knowledge, confidence and belief about alternative solutions

Knowledge and Confidence Knowledge is a measure of the information held by a decision-maker about an alternative. Confidence is a measure of how likely the evaluator believes that an alternative meets a criterion.

Slide 15

"Team Decision-Making Success =" Team Decision-Making Success = Mix of Individual Cognitive Styles + Shared Decision-Making Knowledge + Shared Problem Knowledge + Mix of Alternative Visions + Criteria Understanding Inconsistencies + Criteria Importance Differences + Team Roles + Team Structure + Available Resources

The Relationship between Robust Decisions and Time, Money and other Resources The time, money and resources spent making a decision affects decision robustness. A decision is a commitment of resources. The robustness of the decision-making process is inversely proportional to the time and money it may take downstream to fix the results.

Corporate Decision-Making Styles Decision by running out of time Decision by fiat Decision by coercion Decision by Competition Decision by voting Decision by Inertia Decision by Compromise Decision by Collaboration

Cost for Engineering changes during product development

Slide 20

Envisioning =Understanding the Problem Step 1.   State the issue. Step 2.   Identify the customers. Step 3.   Itemize solution features. Step 4.   Define targets. Step 5.   Measure importance.

Must know issue and criteria for a robust decision

"Step 1:" Step 1: State the Issue. 1/6 1.1 Write a single sentence that describes the issue, question, task, problem statement or area of concern. An issue is the current focus of problem solving requiring the development of new information. A decision made about an issue is generally a call for action dependent on the selection of an option or activity to satisfy some criteria associated with the issue. An issue can be stated as a goal, question, task or problem statement, or area of concern.

Some issues are: Design a front suspension system for a bicycle. What car should I buy? I want to go fast, in comfort. What should I do next on this project? How can we keep the brakes from squealing and still have good deceleration? What are we going to do about Bob?  He seems so disruptive. Find the best employee from the local talent pool for the new marketing position. Where is the best place for our new factory? What is the best Java applet to change the cursor color?

"Step 1:" Step 1: State the Issue. 2/6 1.2 Identify the object, function or process on which activity is focused. Issue = (Object, Function or Process) + Action

"Step 1:" Step 1: State the Issue. 3/6 1.3 Identify the desired action on the object, function or process.

"Step 1:" Step 1: State the Issue. 4/6 1.4 Itemize the initial criteria. 1.5 Capture initial alternative solutions.

"Step 1:" Step 1: State the Issue. 6/6 1.6 Identify the source of the issue. Type 1: Direct Issue decomposition Type 2: Alternative generated issues Type 3: Criteria generated issue

Slide 29

Slide 30

Slide 31

Slide 32

Truths about Criteria If you don’t write down the criteria, they will change during problem solving – in product development this is commonly called “feature creep.” If you don’t know what defines a satisfactory solution, the only way you know you are done is when you run out of time. If you don’t articulate the criteria, different team members will be working toward different goals (trying to meet different criteria) while believing they are working together on the same issue. If you are not careful, you will not make good use of the fact that no two problem-solvers on a team believe the same criteria are important.

Encourage the refinement of criteria.

"Step 3:" Step 3: Itemize the important features of a solution.  1/3 List the features included in the issue statement. Criteria = Feature + Target  + Importance

"Step 3:" Step 3: Itemize the important features of a solution.  2/3 Add features based on those of previous and competitive solutions to similar problems. Refine by listening to voice of the customers Develop and use feature checklists to aid in completeness.

Slide 37

Slide 38

"Step 3:" Step 3: Itemize the important features of a solution.  3/3 Refine the list of features to insure that the criteria are complete, orthogonal, clear, universal, discriminatory, and measurable. Categorize all features as basic, performance or excitement. Insure all features are external.

Kano’s Model of Customer Satisfaction

"Step 4:" Step 4: Define targets for features. 1/2 Identify the basis for evaluation.

"Step 4:" Step 4: Define targets for features. 2/2 Identify the target goal value, units, and type. Insure criterion conditions are known. Refine weak criteria by identifying ways of measuring feature and transforming to strong criteria.

The parts of criterion targets

"Step 5:" Step 5: Measure characteristic importance for each customer. 1/2 Plan methods to honor various customer viewpoints. Measure importance from each customer viewpoint using the fixed sum or ranking method.

Fixed Sum Method  the sum of all the weights must equal a fixed number. 3 -5 times the number of criteria.

Ranking asking customers to weigh criteria is so fraught with error that it is easier and no less accurate  (if more than 6 criteria)just to ask them to rank them. ask the question, " if an alternative scored 0 on all criteria (worst possible alternative) and you could improve just one score from worst to best, which criteria would you choose to improve”? pair-wise fashion by selecting two criteria and asking, “If an alternative could only meet one of these, which criteria would you choose”.

Slide 47

Ideation Risk The risk of not developing good alternatives Step 6. Generate Alternative Solutions. 1/5 Build management structures to encourage the development of many alternative solutions.

"Step 6." Step 6. Generate Alternative Solutions. 2/5 Build working environment to encourage creativity.

"Step 6." Step 6. Generate Alternative Solutions. 3/5 Utilize structured methods to aid in generating alternatives. Morphology, organizing to help generation Brainstorming as a Source of Ideas Using the 6-3-5 Method as a Source of Ideas Using Existing Products and Concepts as Idea Sources, Benchmarking

"Step 6." Step 6. Generate Alternative Solutions. 4/5 Refine alternatives to insure distinctness and variability. Distinct alternatives are readily distinguishable from each other. Variability implies that the alternatives cover the range of potential solutions

Ideation Risk The risk of not developing good alternatives Step 6. Generate Alternative Solutions. 5/5 Monitor the characteristics of new alternatives to provide an alternative filter or criteria amendment. Note new issues generated by alternatives

Evaluation risk Step 7.   Measure level of knowledge. Step 8.   Assess confidence/ determine belief. Step 9.   Determine overall evaluation.

"Step 7:" Step 7: Measure the decision-makers’ knowledge of the alternatives. 1/2 Assess decision-maker’s knowledge about the alternatives’ features.

Slide 55

A Proverb about knowledge

Knowledge scales

Slide 58

Slide 59

"Step 7:" Step 7: Measure the decision-makers’ knowledge of the alternatives. 2/2 Identify methods to increase knowledge and their costs. Analysis either formal or informal Experiments Consultants Vendor representatives Referencing prior documented work Leveraging team knowledge off the knowledge of the individual team members

"Step 8:" Step 8: Determine decision-maker’s belief. 1/3 Assess the decision-makers’ confidence in the alternatives’ ability to meet the criterion target.

Confidence scales

"Step 8:" Step 8: Determine decision-maker’s belief. 2/3 Determine the decision-makers’ belief in the alternatives' ability to meet the criterion target.

Slide 64

"Step 8:" Step 8: Determine decision-maker’s belief. 3/3 Note the evaluation risk in this level of belief.

"Step 9." Step 9. Determine the decision-makers' overall evaluation of the alternatives. 1/3 Estimate each alternative’s total satisfaction. Can be found using iDecision Can be estimated by hand Can be found using the equation programmed in iDecision

Slide 67

"Step 9." Step 9. Determine the decision-makers' overall evaluation of the alternatives. 2/3 Calculate each alternative’s expected value based on all decision-makers evaluation.

Slide 69

"Step 9." Step 9. Determine the decision-makers' overall evaluation of the alternatives. 3/3 Compare the alternative’s expected values from various customer viewpoints.

Slide 71

Slide 72

Slide 73

"Step 10:" Step 10: Decide what to do next. 1/7 Reach agreement and document the result.

Information to help determine what to do next Results of expert knowledge analysis Belief maps Importance weightings Quality and number of alternatives Strength, completeness, orthogonality, clarity, universality, discrimination, and measurability of criteria

The results of the expert knowledge analysis are: a function of the current customer’s importance weighting. a function of the current knowledge indicated by the team. a function of the current confidence indicated by the team. for reevaluation of one criterion at a time.

Strategy for what to do next 1/2

Strategy for what to do next 2/2

Slide 79

"Step 10.2:" Step 10.2: Decide what to do next. 2/7 Further interpret and discuss evaluation information. Based on belief maps. Compare interpretation of information.

"Step 10.3:" Step 10.3: Decide what to do next. 3/7 Develop more evaluation information. Goal is to refine knowledge and then update belief maps.

"Step 10.4:" Step 10.4: Decide what to do next. 4/7 Refine solution features and targets. Return to Steps 3 and 4.

"Step 10.5:" Step 10.5: Decide what to do next. 5/7 Generate new alternative solutions. Return to step 2

"Step 10.6:" Step 10.6: Decide what to do next. 6/7 Negotiate new features, targets or importance weightings. Return to Steps 3 and 4, but with the intent of modifying the criteria and/or opinions about what is important.

"Step 10.7 : Decide..." Step 10.7 : Decide what to do next. 7/7 Decompose the issue into sub issues.

Example using iDecision

Benefits of using the methods 1/4 Encourage sound decision-making skills Organize decision-making to be most effective.  Most decisions are ad hoc; fashioned from whatever is immediately available.  Often this just isn’t good enough. Make robust decisions, decisions that are insensitive to things you can not control. Communicate what is important to other team members.  The largest single problem in teamwork is poor communication.  The methods presented give a framework for decision-making communication.

Benefits of using the methods 2/4 Help the team develop a common understanding of the issue and its alternative solutions. Make meetings more effective.  The methods help structure meetings by developing a strategy and organizing information for easy review. Understand why a decision is not being reached and develop a strategy to resolve the issue. Often a problem is not being resolved and the only action is frustration.  The methods help get problems unstuck.

Benefits of using the methods 3/4 Understand how to get the best out of the people on the team.  Teams are often dominated by a few.  The methods help even the playing field.  Analytically support decision-making regardless of completeness of the problem, the qualitative nature of the evaluation or inconsistency of team member opinions about what is important. Rationally decide what to do next to reach a robust decision.

Benefits of using the methods 4/4 Convince managers that the team has carefully studied the problem and which solution should be implemented. Easily develop documentation of the decision. Reveal the process of decision making for review and reuse.  Understanding and refining the process is important and more easily done if it is structured as developed in this book. Reduce the need to rework the results of non-robust decisions.  In industry this is often referred to as “fire fighting.”  Fire-fighting takes valuable time from working on new issues.

"I have decided to stop..." I have decided to stop now!