Please use this identifier to cite or link to this item:
Title: Stochastic Modelling & Analysis of Dynamic Human-Resource Allocation (StADy)
Authors: Donyina, Adwoa Dansoa
Supervisors: Heckel, Reiko
Fiadeiro, José
Award date: 1-Oct-2011
Presented at: University of Leicester
Abstract: Business processes require involvement of technical components as well as humans to achieve their objectives. However, humans are only predictable to a degree of certainty because, while guided by policies and regulations, they retain the freedom to ignore established procedures or positively react to unforeseen events. Since we cannot change people, we have to be able to recognize their unpredictable behaviour by organising processes in such a way as to benefit from the flexibility of their actions and deal with the problems that arise from it. Business processes tend to be a structured sequence of events; however the assignment of humans to scheduled cases is unstructured. Hence, it is difficult to accurately model and simulate the flexibility of human resource allocation without considering the impact of unpredictable human behaviour. While business processes often have a rigid structure, determining sequences of actions on each individual case, there is flexibility in the selection of cases to be processed as well as in the assignment of human resources. However, such a flexible use of resources poses its own challenges, making process execution difficult to model and predict. In this thesis I propose a methodology and language to support the modelling and evaluation of business process executions with flexible assignment of human resources. The main idea is to model configurations of a business process as graphs and use graph transformation rules in a UML-like syntax to describe the process execution. This model allows to define conditions to temporarily permit actors to exceed their roles in exceptional (escalated) situations, without causing legal repercussions. The evaluation of process execution models is supported by the use of stochastic graph transformations, which allow the qualitative analysis of different organizational policies through simulation. The methodology is presented in four stages of (1) business modelling, (2) process execution design, (3) process encoding and (4) performance evaluation. A case study of a pharmacy process is used to evaluate the approach.
Type: Thesis
Level: Doctoral
Qualification: PhD
Rights: Copyright © the author, 2011.
Appears in Collections:Theses, Dept. of Computer Science
Leicester Theses

Files in This Item:
File Description SizeFormat 
AdwoaDonyinaThesis.pdf6.26 MBAdobe PDFView/Open

Items in LRA are protected by copyright, with all rights reserved, unless otherwise indicated.