Docstoc

Factory Physics

Document Sample
Factory Physics Powered By Docstoc
					                                    TM 792 Special
                                        Topics
                                    Decision Theory
                                        June 23, 2008

                                       Spring 2008
                                Dr. Frank Joseph Matejcik
                                 Ch 15 Scenario Planning
                                 Ch 17 Alternative Decision
                                 Support Systems
                                 Final statements
Frank Matejcik SD School of Mines & Technology   1
          Agenda & New Assignment
    • Tentative Schedule
    • About Final

    • Chapters 15 and 17
    • Decision Analysis for Management
      Judgment 3rd Edition Paul Goodwin &
      George Wright, John Wiley EU
    • Many slides provided by John Wiley.

Frank Matejcik SD School of Mines & Technology   2
                  Access & Overview
               Gone Wednesday - Friday
• Instructor: Dr. Frank J. Matejcik CM 319
   – Work: (605) 394-6066
   – Cell: (605) 431-5731 I’ll try to keep it nearby
   – Home: (605) 342-6871
     Frank.Matejcik@.sdsmt.edu

• Do the book, mostly


Frank Matejcik SD School of Mines & Technology   3
                  Web Resources
       • Class Web site on the HPCnet system
       • http://sdmines.sdsmt.edu/sdsmt/directo
         ry/courses/2008su/tm792M081
       • www.wileyeurope.com/go/goodwin&wright/
       • Streaming video ?
         http://its.sdsmt.edu/Distance/
       • The same class session that is on the DVD
         is on the stream in lower quality.
         http://www.flashget.com/ will allow you to
         capture the stream more readily and review
         the lecture, anywhere you can get your
         computer to run.

Frank Matejcik SD School of Mines & Technology   4
                    Tentative Schedule
        Chapters        Assigned
  12-May 1,2,3          e-mail, contact Discussion Q. 3 page 25
  19-May 3,4,5          Ch 3 1, 3bc, 6 & Ch 4 5, 10
  26-May Holiday
  29-May 5,6,7          Ch 5: 3, 8 & Ch 6: 4, 8 Holiday makeup
  02-June 7,8,9         Ch 7: 1, 3 & Ch 8: 3, 8
  09-June 9,10,11,12    Ch 9: none, Ch 10: 3, Ch 11: 1
  16-June 12,13,14      Ch 12: 5, Ch 13: 9,
  23-June 15,17         No assignment
  30-June Final

Attendance Policy: Help me work with you.
Frank Matejcik SD School of Mines & Technology   5
              Grading for this course
 • I haven’t been able to make up a midterm
   in summer courses.
 • It is a small class so grading hw will be OK
 • Max of (70% Final, 30% hw),
             (30% Final, 70% hw)
 • Final will have a study guide
 • I’ll be handling mailing, too.

Frank Matejcik SD School of Mines & Technology   6
                          Evaluation
  Online survey (evaluation) for the class is open from 6/23/08
  through 7/3/08. The link for the website is below. You can
  only login once, so allow sometime to complete the survey.
  https://theideaonline.org/idea/cs/survey?S=248634/X/F/F/33
  43245432
  If five or more students respond to the survey, I don't have
  to buy a Coke for Stu Kellogg. This will limit Stu's exposure
  to caffeine which will can raise blood pressure, and deplete
  magnesium levels. Also, soft drinks can lead to weight gain.
  So, for Stu's health please complete the survey.
  The dean's secretary, Connie VanBockern, may be able to
  help if there are problems. Connie.Vanbockern@sdsmt.edu
  605-394-5265 (voice)

Frank Matejcik SD School of Mines & Technology   7
                   Exam Study Guide
This will be a problems oriented exam rather than a
  short answer exam, which many of my previous
  classes have been. The later portion of the book was
  oriented to discussion type problems and specialty
  software applications. Accordingly, the exam will be
  have more coverage of the earlier portions of the
  book.
The exam will have questions similar to those at the end
  of chapters 3, 4, 5, 6, 8, and 12. I currently do not
  have the exam written. Exam will be June 30 in class,
  or scheduled afterwards. The IE department secretary
  is off till August, so local students should call or e-mail
  to set an appointment for the exam.
 Frank Matejcik SD School of Mines & Technology   8
             Chapter 15:
          Scenario Planning:
          An Alternative Way
              of Dealing
           with Uncertainty
Frank Matejcik SD School of Mines & Technology   9
             Newsweek Quote 1/28/91
                   page 15
 1. Only most likely discussed
 2. Something like group think is occurring
    (conclusive information ignored)
 3. Contingency planning for low probability
    events was minimal or zero




Frank Matejcik SD School of Mines & Technology 10
                      Visual illusion
       • Figure 15.1
                     Cognitive illusion




Frank Matejcik SD School of Mines & Technology 11
 ‘In times of rapid change, strategic failure is
 often caused by a crisis of perception (that
 is, the inability to see an emergent novel
 reality by being locked inside obsolete
 assumptions), particularly in large, well-run
 companies’
                         Pierre Wack



Frank Matejcik SD School of Mines & Technology 12
      ‘I mistrust isolated trends ... In a period
      of rapid change, strategic planning
      based on straight-line trend
      extrapolation is inherently treacherous
      ... what is needed for planning is …
      multidimensional models that interrelate
      forces – technological, social, political,
      even cultural, along with the
      economies.’

      Alvin Toffler in The Adaptive Corporation
      (1985)
Frank Matejcik SD School of Mines & Technology 13
         Assumptions of scenario planning

     1. Managers are not able to make valid
        assessments of probabilities of unique
        future events

     2. Best guesses of the future may be
        wrong

     3. Minority opinions should be allowed
       ‘airtime’
Frank Matejcik SD School of Mines & Technology 14
                       What are scenarios?

        • A scenario is not a forecast of the future
          – multiple scenarios are pen pictures
          encompassing a range of plausible futures
        • Each scenario has an infinitesimal
          probability of occurrence, but the range
          of the set of scenarios is constructed to
          BOUND uncertainties seen to be inherent in
          the future
        • Unlike probability judgments, scenarios
          highlight the reasoning underlying judgments
          about the future
Frank Matejcik SD School of Mines & Technology 15
                     More on scenarios

 • A major focus is how the future can evolve from
   now to the horizon year

 • Relationship between critical uncertainties,
    predetermined trends and behaviour of actors is
   thought through

 • Decisions are then tested for robustness in
   the ‘wind tunnel’ of the set of scenarios


Frank Matejcik SD School of Mines & Technology 16
           Scenario construction: The extreme
                     world method
 1.   Identify the issue of concern and horizon year
 2.   Identify current trends that have an impact on the issue
      of concern
 3.   Identify critical uncertainties
 4.   Identify whether trends and uncertainties have a
      negative or positive impact on issue of concern
 5.   Create extreme worlds
 6.   Add predetermined trends to both scenarios
 7.   Check for internal consistency
 8.   Add in actions of individuals and organizations



Frank Matejcik SD School of Mines & Technology 17
         Examples of predetermined trends


Demographic:         population growth, birth rates
Technology:          growth rates, production capacity
Political:           power shifts, budget deficits
Cultural:            changing values, spending
                     patterns
Economic:            disposable incomes, investment
                     levels


Frank Matejcik SD School of Mines & Technology 18
                      Business Idea
 •  Systematic linking of the business’s
    competencies and strengths
 1. Competitive Advantage
 2. Distinctive Competencies on which 1 is
    based
 3. The growth mechanism – a positive
    feedback loop


Frank Matejcik SD School of Mines & Technology 19
         An illustrative idea for a business school




Frank Matejcik SD School of Mines & Technology 20
         Testing the robustness of strategies against
                          scenarios




Frank Matejcik SD School of Mines & Technology 21
         An output of the driving forces method




Frank Matejcik SD School of Mines & Technology 22
             Stakeholder structuring space
            missing figure 15.11 figure 15.12




Frank Matejcik SD School of Mines & Technology 23
        When should a company use scenario planning?


   1. When uncertainty is high
   2. Too many costly surprises have occurred in
      the past
   3. Insufficient new opportunities are perceived
      or generated
   4. The industry has experienced significant
      change, or is about to
   5. Strong differences of opinion exist, each of
      which has its merits


Frank Matejcik SD School of Mines & Technology 24
            Typical outcomes of scenario planning


   • ‘This is what we have to do’
       (developing new business opportunities)
   • ‘We better think again’
      (understanding outcomes of plans better)
   • ‘We better watch those dots on the horizon’
      (perceiving weak signals of new
     developments)
   • ‘We are in the right track’
      (moving forward with more confidence)

Frank Matejcik SD School of Mines & Technology 25
           Local council case in the UK
            low probability/high impact
 (i) partner agendas - whether partner
    organizations share the values of the
    council, the commitment to involvement,
    willingness to share resources, etc.,
 (ii) information mapping and understanding
    the basics of the business – how do
    current systems relate to knowledge
    management, can duplicate systems be
    integrated /eliminated, etc.?

Frank Matejcik SD School of Mines & Technology 26
           Local council case in the UK
            low probability/high impact
 (iii) public ownership - is the commitment to
     involvement a solution or an ideology, will the
     public be with the council, how does it relate to
     cultures of youth and the underclasses, will
     participation be hijacked by pressure groups,
     etc.?
 (iv) central agencies as help or hindrance - what is
     the real agenda of central government, does
     system centralization conflict with democracy,
     etc.?
Frank Matejcik SD School of Mines & Technology 27
           Local council case in the UK
            low probability/high impact
 (v) the opportunities and constraints offered
   by new technologies - what resource
   implications are there for the change
   process, what will be the macro-economic
   factors of relevance, how will change be
   managed and what will be the new
   organizational design required to
   implement joined-up government in the
   future
Frank Matejcik SD School of Mines & Technology 28
           Local council case in the UK
                  Four scenarios
 • Forward to the Past – Central government
   runs the show
 • Free Enterprise – customer rules, market
   forces / polarization of society
 • People’s Kailyard – superficial change
 • Technology serves – what we really want
   page 397


Frank Matejcik SD School of Mines & Technology 29
           Council’s Key Implications
 (1) Northshire Council must lead from the
   front, with bold steps in developing an
   integrated and inclusive approach to
   technological innovation. The dangers of
   the small-step and short-term approach
   were highlighted in the kailyard scenario -
   where central government stepped in to
   take control from local government.

Frank Matejcik SD School of Mines & Technology 30
           Council’s Key Implications
 (2) The council must promote democracy in action,
   by making the new technologies serve the
   people and by using technology to develop 'civic
   governance'. They must bring local government
   closer to the community level, developing high
   levels of ability to listen and respond to citizen
   wants and needs. They must develop
   transparency and accountability in their deeds
   and actions, with policies that are meaningful to
   the public.

Frank Matejcik SD School of Mines & Technology 31
           Council’s Key Implications
 (3) New technologies must be used to
   demonstrate the competence of local
   government, achieving public confidence
   and support through the provision of
   responsive, community-oriented services,
   more customized services, while at the
   same time applying the technologies to
   support inclusion and to reduce
   inequalities.
Frank Matejcik SD School of Mines & Technology 32
           Council’s Key Implications
(4) Northshire Council must use the new
  technologies in order to promote itself as the
  'home for sustainable value creation'.
(5) The council must proactively promote and
  lobby for settlement of the subsidiarity
  debate in favor of governance at the local
  level.
(6) Finally, in developing short-term solutions
  to immediate problems, the council must
  watch out that long-term aspirations remain
  the guiding light.
Frank Matejcik SD School of Mines & Technology 33
             Benefits from scenarios
 (i) shared insights for participants in the
    process,
 (ii) alternative ways forward tested against
    scenarios,
 (iii) motivation for action from these insights,
 (iv) agreement on a well-defined way
    forward,
 (v) agreement on technology choice to
    support the way forward.
Frank Matejcik SD School of Mines & Technology 34
                Table 15.1 page 401




Frank Matejcik SD School of Mines & Technology 35
           Combine Scenario Planning
             and Decision Analysis
 • A procedure is given which uses SMART
   analysis
 • An example is included
 • Designing new strategies appears to be a
   a more common approach




Frank Matejcik SD School of Mines & Technology 36
     Chapter 17: Alternative
       Decision-Support
           Systems
           Summary
Frank Matejcik SD School of Mines & Technology 37
                    Two Approaches
 • Linear Models – Build a statistical model of
   the a person’s decisions and use the
   statistical model instead of the person

 • Expert System – Build a computer model
   of the decision process



Frank Matejcik SD School of Mines & Technology 38
             What is an expert system? AI

    ‘The modelling, within a computer, of expert
      knowledge in a given domain, such that the
      resulting system can offer intelligent advise or
      take intelligent decisions’
                                British Computer Society

    • The system should be able to justify the logic
      and reasoning underlying its advice or
      decision making.


Frank Matejcik SD School of Mines & Technology 39
                      Expert knowledge

 • Knowledge elicitation – gaining knowledge
   from an expert
 • This knowledge may consist of many
   unwritten rules of thumb
 • Knowledge bottleneck – difficulty of
   eliciting expert knowledge and transferring
   it to a computer program


Frank Matejcik SD School of Mines & Technology 40
                 Representing knowledge in
                      expert systems
   Production rules:
      IF (condition in database) THEN (action to update the
      database)
   Control structure or inference engine:
     determines what rule is to be tried next
   Forward chaining:
      following pathways through from known facts to
      resulting conclusions
   Backward chaining:
     choosing hypothetical conclusions and testing to see
      if the necessary rules underlying the conclusions hold
      true
Frank Matejcik SD School of Mines & Technology 41
 •     Successful systems must be able to
       interface effectively with their users in
       order to:
     1. Gain the information needed to test the
         rules
     2. Give understandable advice in plain
        English and justify the logic and
        reasoning underpinning the advice
       given or decision made


Frank Matejcik SD School of Mines & Technology 42
        Indicators of whether an expert
          system can be built within a
             reasonable time frame
 1. That the subject domain has been
    formalized, e.g. do manuals exist?
 2. That the subject domain is amenable to
    verbal expression, e.g. discuss on the
    phone
                              Wright and Ayton (1987)


Frank Matejcik SD School of Mines & Technology 43
                   Underwriting options figure 17.1




Frank Matejcik SD School of Mines & Technology 44
         Example of rule base of the geographical module
                             fig 17.2




Frank Matejcik SD School of Mines & Technology 45
            Statistical models of judgment

   1. Obtain data on the judgments an
      expert made and the variables he/she
      used to make the decision.
     E.g. an expert judges risk of firms failing:
         p(a business      Financial    Financial    Financial
              will fail)     ratio 1      ratio 2      ratio 3
                  32%            0.8          1.2          0.6
                  51%            1.2          0.9          0.2
                      ..           ..           ..           ..
                      ..           ..           ..         etc

Frank Matejcik SD School of Mines & Technology 46
            Statistical models of judgment

 2. Fit a statistical (regression) model to the
   data to capture the weight the expert
   placed on each variable.
 E.g.
 p(a business will fail)
 = 0.3 Financial ratio1 + 0.1 Financial ratio2
    + 0.05 Financial ratio3


Frank Matejcik SD School of Mines & Technology 47
            Statistical models of judgment

 3. Use the model of the expert, rather than
   the expert him- or herself, to make
   predictions

    –this model is likely to outperform the
    expert because it ‘averages out’
    inconsistencies in the expert’s judgment
    while retaining the ‘core’ of his, or her,
    expertise.
Frank Matejcik SD School of Mines & Technology 48
   Example of
    statistical
   modeling of
    judgment




Frank Matejcik SD School of Mines & Technology 49
                    Final Words of Advice

 • Are the my assumptions about the
   business environment and the future
   valid? Decision Framing and Scenarios
 • Who should I involve in the decision?
 • Have I made an adequate search for
   alternative courses of action
 • How much effort is the decision worth?
 • Table 17.1 Summary of Book’s techniques
Frank Matejcik SD School of Mines & Technology 50
                          Final Slide
 •   Just an exam next week
 •   Study guide may have more details
 •   I will try to grade things before the exam
 •   Keep Stu healthy; Fill out the Online
     survey




Frank Matejcik SD School of Mines & Technology 51

				
DOCUMENT INFO
Shared By:
Categories:
Tags:
Stats:
views:2
posted:12/13/2011
language:
pages:51