Premium
A COMPARISON OF ALTERNATIVE PRODUCTION MANAGEMENT COEFFICIENT DECISION RULES
Author(s) -
Hamner W. Clay,
Carter Phillip L.
Publication year - 1975
Publication title -
decision sciences
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.238
H-Index - 108
eISSN - 1540-5915
pISSN - 0011-7315
DOI - 10.1111/j.1540-5915.1975.tb01024.x
Subject(s) - miller , variance (accounting) , production (economics) , computer science , operations research , work (physics) , representation (politics) , regression analysis , decision rule , linear regression , management science , mathematics , economics , artificial intelligence , machine learning , microeconomics , engineering , mechanical engineering , ecology , accounting , politics , political science , law , biology
Production managers are called on to combine cues to arrive at work force and production level decisions. Bowman has developed a theory which claims that a manager makes good decisions on the average, but that he may exhibit a high degree of variance in his behavior. Mathematical representation of such managers can be constructed to capture critical aspects of their judgmental strategies. This paper examines four different linear regression models in an attempt to specify which model is best for prescription of decision making behavior in an organization. The models include (1) the Typical Manager Model, (2) the Best Manager Model, (3) the Composite Manager Model, and (4) the Best Composite Manager Model. The study described in this paper concerns an experiment conducted by Moskowitz and Miller in 1972. Eighty‐seven graduate students in industrial management were randomly assigned to one of six information groups. Each student made a series of production and work force decisions to minimize total production, work force, and inventory‐related costs. Moskowitz and Miller concluded that Bowman's Typical Manager Model was a better model than the managers' actual initial plans. While there are no disagreements with this finding, a reanalysis of this data was needed to see if perhaps another mathematical rule might prove even better than the Typical Manager Model recommended by Bowman. This reanalysis of the Moskowitz and Miller results enables us to point up the strengths and weaknesses of using regression models in managerial decision making tasks. Suggestions for future research are discussed in detail.