Modeling Organizationalg g Architecture ESD 342 Class 20 April 22, 2010p , Christopher L. Magee © 2007 Chris Magee, Engineering Systems Division, Massachusetts Institute of TechnologyLearning objectives Appreciate some additional range of organizational research including some simple organizational research including some simple models Appreciate one organizational modeling approach relative to our growing understanding of the use of network models The emphasis is on ways of thinking/modeling- The emphasis is on ways of thinking/modeling my management experience nonetheless informs my perspectives. I do not consider the very important issue of culture dealt with by JM © 2007 Chris Magee, Engineering Systems Division, Massachusetts Institute of TechnologyLectures 19/20: Outline A brief tour of research on organizations The organizational design problem The organizational design problem Design variables, fundamental metrics and the bottom line Processes Properties Properties Organizational Design/Architectural Analysis by selected, simple quantitative models and a “modeling framework”. Arrow; Sah and Stiglitz Arrow; Sah and Stiglitz Simple decision-making non-network models Dodds, Watts and Sabel Network model incorporating hierarchy as base Network model incorporating hierarchy as base Information transfer for problem solving Robustness assessments and identification of superior structure Assessment of the contribution of DWS paper © 2007 Chris Magee, Engineering Systems Division, Massachusetts Institute of Technology Possible future work and ConclusionsModeling Organizational issuesModeling Organizational issues After the introduction about organizational design in L19 several aspects of modeling that relate to L19, several aspects of modeling that relate to organizational structure (or architecture) are now briefly explored: Decision Theory Communication Communication Note that both of these are properties models and Note that both of these are properties models and do not discuss or try to look at models for formation or evolution of actual organizational structure or the development of rules etc © 2007 Chris Magee, Engineering Systems Division, Massachusetts Institute of Technology structure or the development of rules, etc.Modeling Decision Making Processes Items to be covered (briefly) l l d An axiom concerning multiple decision makers (team or organizations) with multiple alternatives (Arrow) Multiple decision makers and decision structures/organizations (Sah & Stiglitz) © 2007 Chris Magee, Engineering Systems Division, Massachusetts Institute of TechnologyArrow’s Impossibility TheoremArrow s Impossibility Theorem IndividualIPreferencesA>B> C, A> CA vs. BAB vs. CBA vs. CAII B>C> A, B> A B B C,III C>A> B, C> B A C CGroup preferences A>B B>C C>AGroups using majority ruleare not necessarily transitive© 2007 Chris Magee, Engineering Systems Division, Massachusetts Institute of TechnologyImplications to Consensus Decisions, Cooperation and OrganizationsCooperation and Organizations There is a real difficulty with intransitivity-almost certain in a large group with a large number of options-h li l h i k Large teams with multiple choices to rank are unproductive and should be avoided Single person decision after all team input is heard is one possible alternativepossible alternative Avoid group participation in ranking-just binary decisions or picking the best among a group of alternatives Sum ranking votes on single alternatives (with an arbitrary tie- Sum ranking votes on single alternatives (with an arbitrary tie breaking rule and/or elimination of lowest total) Facilitators (and/or decision-makers) can force a series of two way choices and eliminate any losers from further consideration Some organizational hierarchy is essential to effectiveness particularly if one wants to rank a long © 2007 Chris Magee, Engineering Systems Division, Massachusetts Institute of Technology p y g list of attributesDecision-making Structures and Organizational ImplicationsOrganizational Implications Sah and Stiglitz work is foundation and expanded by Catalani and Clerico Models for different decision-making structures Framework involves “approving (or not) generalized Framework involves approving (or not) generalized Projects” Good (g) or bad (b) projects can be accepted (a) by individuals with probabilities p and pindividuals with probabilities p1 and p2 © 2007 Chris Magee, Engineering Systems Division, Massachusetts Institute of Technology“good” “bad” project or project or didesign didesign change change AcceptAccept P1 P1 P2 P2 Ideal = 1.0 Ideal = 0.0 T II Type II errors Reject 1-p1 Type I error Type I error 1-p2 Amount of Type I and II Errors for © 2007 Chris Magee, Engineering Systems Division, Massachusetts Institute of Technology Amount of Type I and II Errors for individual decision-makert tDecision-making Structures and Organizational Implications IIOrganizational Implications II Sah and Stiglitz work is foundation and expanded by Catalani and Clericoy Models for different decision-making structures Framework involves “approving (or not) Projects” G d ( ) b d (b) j b d ( ) b Good (g) or bad (b) projects can be accepted (a) by individuals with probabilities p1 and p2 for “polyarchy”-simultaneous judgment and any one person acceptance Pa g = p1(2-p1) and Pa b = p2(2-p2) for “hierarchy”-series of decisions with only approved considered at next level, Pa g = p1 2 and Pa b = p2 2 © 2007 Chris Magee, Engineering Systems Division, Massachusetts Institute of Technologyt tDecision-making Structures 2 Decision making Structures 2 Hierarchy of n people rarely acceptsy p p y p anything bad (but often rejects good changes- type II increases) “and gate” analogy and redundancy Polyarchy of n people rarely rejects thi d (b ft b danything good (but often accepts bad changes- type I increases) “or gate” analogy or gate analogy © 2007 Chris Magee, Engineering Systems Division, Massachusetts Institute of TechnologyGeneralization to “Committees”Generalization to Committees Can vary number of people on committee, n and number who must approve for acceptance, v. Optimum
View Full Document
Unlocking...