A decision making model for management executive planned

Document Sample
A decision making model for management executive planned Powered By Docstoc
					A decision making model for
management executive planned
behaviour in higher education


Laurentiu David M.Sc.Eng., M.Eng., M.B.A.

Doctoral student at the Ontario Institute for Studies in Education
University of Toronto, CANADA
¢   Statement of purpose
¢   Introduction
¢   Brief literature review
¢   Linear programming model
¢   Method
¢   Case Study
¢   Research model
¢   Theory of planned behaviour
¢   Competing Value Framework
¢   Mathematical application of the model
¢   Data
¢   Conclusion
¢   The End
Statement of purpose

¢   The purpose of this paper is two fold:
    l to develop a linear programming
      solution for decision making
    l to offer a possible justification for the
      existing gap between an agent
      intention to pursue a particular
      behaviour and the actual behaviour
¢   In the past, the decision making processes were based on
    intuition, experience and/or a mix of the two.
¢   Even today, in spite of the development of a large array of
    mathematical planning methods, business – planning
    processes include subjective judgements making decisions
    frequently vague.
¢   As a result, it can be claimed that since a decision can be
    vague it can be represented on fuzzy numbers.
¢   Dyson (1980, p.264) purported that fuzzy programming
    models should not be seen as a new contribution to multiple
    objective decision making methods, but rather as a lead to
    new conventional decision methods.
¢   The present paper builds its structure on an existing linear
    programming technique developed by Li and Yang (2004,
    p.271) that takes into consideration a multidimensional
    analysis of preferences in multiattribute group decision
    making under fuzzy environments.
Brief Literature Review
¢   According to Treadwell (1995, p.93) who claimed that” the
    dialogue between the human sciences and fuzzy set theory
    has been scattered, unsystematic, and slow to develop” the
    fuzzy set is not the panacea for dealing with the world of
    uncertainty in certain terms, but it is a strong contender.
¢   Smithson (1987, p.11) noted the fact that the principal value
    he found in fuzzy set theory is that it generates alternatives
    to traditional methods and approaches, thereby widening the
    range of choices available to researchers.
¢   According to Lazarevic and Abraham (2004, p.1) decision
    processes with multiple criteria are dealing with human
¢   The human judgement element is in the area of preferences
    defined by the decision maker (Chankong, & Haimes, 1983).
¢   Kaufmann and Gupta (1998, p.7) considered that classical
    social system models are suited for simple and isolated
    natural phenomena.
Linear programming model
¢   1. Evaluate the parameters of the decision maker
¢   2. Determine the decision maker’s order of preferences
¢   3. Determine the linguistic ratings of the variables (roles)
¢   4. Map the decision maker opinion using the linguistic rating
    for each of the variables (roles) under each attribute
¢   5. Construct the fuzzy decision matrix and normalize the
    positive trapezium fuzzy number decision matrix
¢   6. Construct the linear programming formulation
¢   7. Solve the system of equations
¢   8. Obtain the weights vectors and the fuzzy positive ideal
¢   9. Calculate the distance of each variable (role)
¢   10. The determine the ranking order of each variable (role)
Case Study

¢ Place: Higher education institution
¢ Position: Management

¢ Decision Maker(s): 1

¢ Assumptions:
     •   Research model
     •   Theory of Planned Behavior
     •   Competing Value Framework
     •   RREEMM
Research Model
Theory of Planned Behaviour
¢   The original linear formulation of the theory of planned behavior in its
    simplest form is expressed by the following mathematical function:

¢   BI – behavioral intention,
¢   AB – attitude toward behavior
¢   b – the strength of each belief
¢   e – the evaluation of the outcome or attribute
¢   SN – social norm
¢   n- the strength of each normative belief
¢   m – the motivation to comply with the referent
¢   PBC – perceived behavioral control
¢   c- the strength of each control belief
¢   p – the perceived power of the control factor
¢   W – empirically derived weights
Competing Value Framework

subject to:
¢   A linear programming problem was developed once the data from
    the trapezium fuzzy number matrix was introduced into the new set
    of equations.
¢   The objective function was then configured to be as it follows:
¢   The objective function was subjected to a set of over 20 equations
    containing over 30 distinct variables.
¢   max
¢   Because of the lengthy aspect of equations the mathematical
    calculus has been omitted from the paper.
¢   Solving the linear equations using the Simplex method helped with
    obtaining the and vectors.
¢   The ranking order of the possible roles was obtained by calculating
    the distances from the generated fuzzy positive ideal solution.
¢   The generated ranking order places R4 – director role at the best
    choice as the outcome maximizing executive behaviour when it
    comes to offer a solution to the low enrolment situation since:
¢   The numerical example showed the fact that even
    though the actor had a particular pre-action set of
    roles’ preferences when it came to solve a particular
    problem the final role choice differed from the
    expected one.
¢   The gap between the agents’ intentions and
    behaviours can be even better exemplified when the
    number of actors is increased.
¢   In the envoi, the biggest assumption of the model is
    that the agent final choice will correspond to the
    mathematical solution found by employing the herein
    proposed fuzzy logic anchored method of calculus.

¢   Thanking you for attending this
    presentation I am inviting you in the
    next 10 minutes to address your

Shared By: