Process improvement : the creation and evaluation of process alternatives

M. Netjes

Research output: ThesisPhd Thesis 1 (Research TU/e / Graduation TU/e)

1354 Downloads (Pure)

Abstract

Companies continuously strive to improve their processes to increase productivity and delivered quality against lower costs. With Business Process Redesign (BPR) projects such improvement goals can be achieved. BPR involves the restructuring of business processes, stimulated by the application of information technology. Although BPR is widely applied in industry, a systematic approach that helps a team in deriving designs for better performing business processes is lacking. The approach for Process Improvement by Creating and Evaluating process alternatives (in short: the PrICE approach) is developed to describe and support the concrete steps that will lead a design team from the as-is process to the to-be process. The starting point for the PrICE approach is a model of an existing process. The as-is model contains tasks and their execution ordering, data elements that are created and used, resources and their allocation and performance information. First, redesign opportunities are identified in the process model. A redesign opportunity leads to a combination of a certain redesign operation and a process part on which this operation can be applied. The PrICE approach consists of four steps. The first step of the PrICE approach describes the selection of redesign operations. Eight redesign operations have been developed, each supporting a particular type of redesign creation. We introduce two possible means to select applicable operations: process measures and process mining. Process measures provide a global view on the characteristics of the process and their values may reveal weaknesses in the process. The idea of process mining is to discover, monitor and improve business processes by extracting knowledge from event logs. Process mining results point out what type of changes may be beneficial. Moreover, bottlenecks, i.e., parts of the process that need improvement, are detected. The second step of the approach is the selection of process parts. In this step we focus on the selection of a process part in such a way that the application of a redesign operation results in a correct process model. The third step of the approach concentrates on the creation of process alternatives. An application of a redesign operation to a selected process part results in an alternative process model. An overview of the created alternatives is provided with the process alternatives tree. The root node of the tree is the original model and the other nodes represent the created alternatives. Each of the nodes may serve as a starting point for the creation of another alternative. In the last step of the approach, the performance of the process alternatives is evaluated with simulation. Simulation provides quantitative estimates for the performance, e.g., on time or costs, of a process model. By comparing the simulation results of the models in an alternatives tree, a quantitatively supported choice for the best alternative model, the to-be process, can be made. The PrICE approach is supported with the PrICE tool kit. The tool support is implemented as part of the Process Mining (ProM) framework. The tool kit supports the application of the various steps of the approach. The first two steps are supported with the process mining techniques that are available in ProM. The main features of the PrICE tool kit are the user guidance in the selection of process parts, the creation of process alternatives, the construction of the process alternatives tree and the evaluation of the alternatives with simulation. After the selection of a redesign operation, a process part for redesign is selected by the user by clicking on the tasks in the process model. Colors are used to guide the user and show which tasks may be added to the current selection to form a process part. This way, it is ensured that the input for the creation of a process alternative is such that a correct alternative model can be created. After the creation of an alternative model, the process alternatives tree is updated with a new node representing this alternative. Each node in the alternatives tree can be selected as starting point for the creation of another process alternative. With regard to the evaluation of the alternatives, one can select a subset of nodes for simulation or simulate the complete tree. A simulation study is performed in batch, i.e., all selected models are simulated without user interaction. Afterwards, the simulation results are displayed on the tree nodes. In addition, colors are used to guide the user in finding the best performing alternatives. The developed tool support demonstrates the feasibility of our ideas. This feasibility is also illustrated with several applications of the tool kit to real life processes. Apart from the development of the PrICE approach and tool kit, the thesis includes several other contributions. A contribution is the creation of correct process models. We refer to a process model as correct if the workflow structure is sound and if the data distribution is correct. A correct data distribution is an assignment of the data elements to the tasks in the process in such a way that the data elements necessary for the execution of a task have been written when the task becomes enabled. Requirements on the workflow structure and data distribution are set on the selection of process parts and the creation of alternatives to ensure the construction of correct process alternatives. Another contribution is the overview of the created process alternatives with the process alternatives tree. An alternative model may be created from the original model (the root node) or from one of the alternative models (any other node). The alternatives tree is also used as input for the evaluation of the performance of the alternatives and to provide an overview of the simulation results. A final contribution is the enhancement of the practical use of simulation for process redesign. On the one hand, the automation of the simulation study reduces the necessary time investment because intermediate input from the user is not required. On the other hand, we present a simulation plan that facilitates the understanding of the various aspects that should be addressed in a simulation study.
Original languageEnglish
QualificationDoctor of Philosophy
Awarding Institution
  • Industrial Engineering and Innovation Sciences
Supervisors/Advisors
  • van der Aalst, Wil M.P., Promotor
  • Reijers, Hajo A., Copromotor
Award date28 Sep 2010
Place of PublicationEindhoven
Publisher
Print ISBNs978-90-386-2320-7
DOIs
Publication statusPublished - 2010

Bibliographical note

Proefschrift.

Fingerprint Dive into the research topics of 'Process improvement : the creation and evaluation of process alternatives'. Together they form a unique fingerprint.

Cite this