An Evolutionary Approach for Project Organization Design: Producing Human-Competitive Results using Genetic Programming

Created by W.Langdon from gp-bibliography.bib Revision:1.3872

@PhdThesis{KHosraviani_2005,
  author =       "Bijan KHosraviani",
  title =        "An Evolutionary Approach for Project Organization
                 Design: Producing Human-Competitive Results using
                 Genetic Programming",
  school =       "Department of Civil and Environmental Engineering,
                 Stanford",
  year =         "2005",
  month =        dec,
  note =         "submitted as a doctoral dissertation to Stanford
                 University",
  keywords =     "genetic algorithms, genetic programming",
  URL =          "http://crgp.stanford.edu/publications/dissertations/KHosraviani_2005.pdf",
  size =         "141 pages",
  abstract =     "In the complex and rapidly changing business
                 environment of the early 21st century, designing an
                 effective and optimised organisation for a major
                 project is a daunting challenge. Project managers have
                 to rely on their experience and/or trial and error to
                 come up with organisational designs that fit their
                 particular projects. Painful and costly experience in a
                 wide range of governmental and private organisations
                 has demonstrated that projects to develop buildings,
                 software and other products often fail, not because the
                 design of individual components was at fault, but
                 rather because the organisation performing the complex
                 supervision and coordination tasks required for system
                 integration failed due to information overload.

                 The Virtual Design Team (VDT) simulation system, based
                 on the information processing theories of organization
                 science, was a successful attempt to develop an
                 analysis tool for project organization design (Jin and
                 Levitt, 1996). However, like the analysis tools that
                 support many other design processes, VDT has no
                 inherent ability to improve or optimize current designs
                 automatically. It simply predicts performance outcomes
                 ? in terms of time, cost and several measures of
                 process quality ? for a particular project organization
                 design alternative. A VDT user must thus experiment in
                 {"}What if?{"} mode with different design alternatives
                 in an attempt to find better solutions that can
                 mitigate the identified risks for a given project
                 configuration. The problem has many degrees of freedom,
                 so the search space for better solutions is vast.
                 Exploring this space manually is infeasible. VDT relies
                 on the expertise of the human user, guided simply by
                 intuition about ways to improve on prior designs, to
                 find better solutions. So it offers no guarantee of
                 optimality.",
  abstract =     "Our research extends the capabilities of VDT and other
                 similar organizational analysis tools by going beyond
                 computer support for {"}What-if?{"} analysis to
                 automated design of project organizations. Evolutionary
                 computing methods such as genetic programming are used
                 to design and develop a postprocessor for VDT to help
                 project managers find near-optimal designs for their
                 project organisations.

                 This dissertation describes in detail the approach I
                 developed to represent project organisation design
                 alternatives in a genetic programming format, so that
                 the design can effectively evolve. In addition, it
                 demonstrates how I was able to represent different
                 project performance objectives and constraints in a
                 fitness function which can successfully guide the model
                 toward searching for better designs.

                 A preliminary version of my postprocessor optimiser
                 beats the best human trial-and-error solutions
                 developed by more than 40 teams over the past eight
                 years. The postprocessor was awarded a Silver Medal for
                 human-competitive results in genetic and evolutionary
                 computation at the GECCO-2004 Conference.",
  abstract =     "I discuss why I chose the evolutionary computing
                 approach as opposed to classical optimisation
                 methodology, and show some of the advantages and
                 limitations of the evolutionary approach. There was no
                 formal theory of project organisation design nor any
                 analysis tools for predicting the performance of
                 project organisations prior to the development of VDT
                 in the 1990s. Not surprisingly, therefore, research by
                 C.B. Tatum in the early 1980s found that current
                 human-developed project organisation structures are the
                 result of {"}natural{"} trial-and-error evolutionary
                 processes. By applying the evolutionary computing
                 approach to organisation design, my model is thus
                 actually mimicking the nature of human organisation
                 design. In addition, I demonstrate how my approach can
                 create a powerful Human-Computer Interaction (HCI)
                 environment that can motivate humans to think
                 {"}outside the box{"} when designing project
                 organisations.

                 Using a combination of {"}intellective{"} (theorem
                 proving) and {"}emulation{"} (natural, empirical)
                 experiments, I validate the postprocessor's
                 {"}near-optimal{"} solutions against findings of
                 organisational contingency theory and human-derived
                 solutions for a set of real test cases. By showing that
                 {"}optimal{"} structure depends on the relative
                 emphasis of time, cost and process quality outcome
                 metrics, I extend contingency theory to develop a
                 richer {"}micro-contingency theory{"} for project
                 organisations.

                 This research represents a significant step towards
                 closing the relevance gap between organisation theory
                 and organisation practice by addressing the issues of
                 organisational design prescriptively. I analyse
                 alternatives in terms of fitness functions that
                 evaluate specific designs for {"}survival{"} and
                 {"}reproduction{"} in the spirit of contingency theory.
                 Finally, the thesis concludes with a summary of the
                 contributions of this research in the three areas of
                 organisation science, project management, and computer
                 science.",
}

Genetic Programming entries for Bijan KHosraviani

Citations