A socio-technical approach to business process simulation

Abstract

This paper describes a socio-technical approach to business process redesign through the investigation of complex interactions and dependencies among humans and IT systems of organisations. The focus is on the need to assess business process performance early in the redesign process, to prevent organisational failures. The method is based on human performance quantification and is supported by a tool that enables scenario-based evaluation of prospective organisational processes through simulation. The approach combines probabilistic and subjective measures of tasks and communication acts in business processes to quantify business performance in terms of cycle time. The approach models business processes as a set of scenarios of sequential activities where the dependencies between actors, IT systems and tasks are explicitly defined. Business process performance assessment is achieved through a systematic walkthrough of the process model using these scenarios. Human performance constitutes an important parameter to business process performance and is modelled based on Human Performance Shaping Factors (PSF) and assessed using Bayesian Belief Networks (BBN). Process cycle time is calculated using aggregates of task and communication completion times, and calibrated using performance estimates of each of the agents in the scenario. The method enables trade-off analysis among candidate process models and identification of performance bottlenecks early in the design phase. A radiology process improvement case study is presented that demonstrates the use of the method and the tool.