PhD Students

The models developed offer a valuable contribution for reliable predictions of measles outbreaks with the aim toward achieving optimal and/or improved control strategies. A detailed account of the application of Operational Research (OR) in social contact networks epidemiology is provided to expand the theoretical discussion on the use of this approach for early detection of cases and timely interventions, in areas with health system constraints. Results of the SEIR simulation model of measles spread strongly indicate how the spread of measles largely depend on the social contact network rates among infected individuals within a population of the rural communities. This offers prospects for developing and proposing optimal and/or improved control strategies for measles. The performed sensitivity analysis on selected parameters of the SEIR simulation model of measles spread indicates that measles epidemics with higher mean infectious periods may tend to peak earlier compared to measles epidemics with a lower mean infectious period. Findings of the SEIR vaccination model indicate a need for targeted vaccination for individuals of 6 months to 19 years of age and children of 6 months to 15 years of age respectively, but equally for older age groups who were born before or missed the second dose schedule, given the deficiencies in health infrastructure that hinder effective mass vaccination programmes. The research contributes theoretically and methodologically to existing applications of social contact networks modelling for airborne infectious diseases in areas with health system constraints. It sets out implications for the design of effective vaccination programmes for control of measles in Tanzania and highlights areas for further research.

The main thesis of this dissertation is that the Structured MCDM methodology not only works in a participatory context but, also, adds to and enhances participation.  It was applied in new geographical area, culture, language and level of development. This study demonstrates that the “Direct” and “Interactive” nature of the methodology can be used to guide the roles of the Participation Facilitator or Decision Advisor (DA) and help to shape the stages of the participatory process. The study also resolves the previous shortcoming of the Structured MCDM Methodology, that it did not take account of the presence of parties other than the DA and DM; particularly the Sector Experts (SE) and the Rule Makers (RMs) This study adds to the emerging research and debate on participatory process design, implementation and evaluation. It demonstrates that the Structured MCDM methodology follows an adapting Conflict Resolution process and, also, suggests that effective participatory processes should follow these stages: trust building, empowerment, creating sense of ownership and commitment.

The empirical case has shown that where contextual issues are complicated, they tend to take a bigger role in the discussions especially when the technical aspects seem to be simple. This dissertation has demonstrated how the DA could handle the contextual aspects of the problem by magnifying certain aspects without contradicting the methodology or Nomology principles on which it is based.

Lastly, Nomology is, at the highest level, about a balance between the subjective “ologies” and the objective “onomics”, and about a balance between “self” and “other”.  This dissertation has extended this latter aspect, showing that it can take account of a variety of “self-to-other” relationships.

Michael Patrick Phelan, B.Sc., M.Mangt.Sc. “Evolving multi-echelon supply chain inventory and ordering policies using biologically inspired grammar based meta-heuristics”

In recent years there has been a growing literature in the area of biologically inspired algorithms, particularly genetic algorithms and genetic programming and their applications to supply chain modelling and inventory control optimisation. Due to the rigidity of the genetic algorithms approach, it is difficult to change the underlying model logic and consequently difficult to add richness to the supply chain. While the application of genetic programming provides a more flexible approach than that provided by genetic algorithms, to date its application has been limited to small supply chain modelling problems in relation to optimal inventory policies. This research introduces Grammatical Evolution, a relatively new biologically inspired algorithm in computer science to the field of supply chain optimisation, employing human readable rules called grammars. These grammars provide a single mechanism to describe a variety of complex structures and can incorporate the domain knowledge of the practitioner to bias the algorithm towards regions of the search space containing better solutions.

The primary research question of this work asks if grammatical evolution can provide managerial insights and cost effective heuristics for supply chain optimisation across a range of realistic scenarios. The methodology used in this research is experimental. Given the stochastic nature of simulating supply chain models with stochastic demand; a statistical analysis of several runs is employed to evaluate the cost effectiveness of supply chain ordering policies generated by grammatical evolution. The supply chains modelled incorporate more realistic features including: inventory allocation policies, payment incentives, linear and distribution supply chain structures, fast and slow moving stochastic demand and capacity constraints on the warehouse and logistics. Using different grammars, ordering policies that minimise the costs in centralised supply chains are compared to policies that balance the associated risks and costs across the supply chain partners in decentralised supply chains.

On the experimental evidence obtained across an extensive range of scenarios, this research demonstrates the flexibility of grammatical evolution as a supply chain optimisation tool and also its ability to adapt to the objective of each scenario, delivering cost effective ordering policies. The grammars incorporating domain knowledge consistently generate the best supply chain ordering policies. Combining this powerful optimisation approach with more realistic models and incorporating their own domain knowledge, practitioners can develop grammars to bias the grammatical evolution algorithm towards funding better supply chain ordering policies. The experiments in this work demonstrate that grammatical evolution can deliver a range of solutions for the same problem, enabling practitioners to compare and contrast policies, highlighting questions that impact on the underlying supply chain strategy. However, it is left up to the expertise of the supply chain practitioner to analyse the managerial implications of these policies.