Cost-Informed Log Analysis and Process Improvement
Organisations are constantly seeking new ways to improve operational efficiencies. This research study investigates a novel way to identify potential efficiency gains in business operations by observing how business operations are carried out in the past and then exploring better ways of doing things from a cost perspective. As trade-offs between time, cost and resource utilisation are typically considered during process improvement activities, this paper demonstrates how such trade-offs can be incorporated in the assessment of alternative process execution scenarios by means of a generic cost structure. A cost optimisation environment has been defined as a first step towards concretisation of a cost-informed process improvement. A hybrid genetic algorithm-based approach has been proposed to explore and assess alternative process execution scenarios. Experiments were conducted with different variants of the genetic algorithm to evaluate the feasibility of the approach. The findings demonstrated that a hybrid genetic algorithm-based approach is able to make use of cost reduction as a way to identify improved execution scenarios that are better in terms of reduced case durations and increased resource utilisation. The ultimate aim is to utilise cost-related insights gained from such improved scenarios to put forward recommendations for reducing process-related costs within organisations.
- XES Files:
- XML Files:
- Source Code:
The previous work takes the original execution history of a business process (event log) as an input, heuristically introduces cost-reducing perturbations, and produces an alternative execution scenario (perturbed event log). Often the cost of the alternative execution scenario is significantly lower compared to the original execution scenario. Analysts will want to know what have exactly changed, or what are the differences between the two execution scenarios. Identifying the differences between the original event log and the perturbed event log could facilitate organisations in gaining process improvement-related insights. However, there is a lack of automated techniques to detect the differences between two event logs. Therefore, this research aims to develop visualisation techniques to provide targeted analysis in terms of resource reallocations and activity rescheduling. The differences between two event logs that process analysts would want to know are first identified. Appropriate visualisation types, as well as design principles that will best portray the changes between the two event logs, are conceptualised and realised with a number of visualisations. With the proposed visualisations, process analysts will then be able to identify the changes that have been made from the resource and timing perspectives that resulted in a more efficient business process. Ultimately, process analysts can make use of this comparative information to initiate evidence-based BPI activities.
- XES Files:
- User Evaluation:
- Source Code:
To provide additional support for process analysts with their decision making, the next step is to pinpoint specific process-related changes, as well as to determine their impact on execution cost. Utilising the previously proposed hybrid genetic algorithm, a collection of alternative execution scenarios were generated. Next, changes related to resource reallocations and activity rescheduling were extracted by identifying the differences between the original and alternative execution scenarios. The overall costs of the alternative execution scenarios were computed as well. After that, the value of the features was transformed to take into account the magnitude of their changes. As a result, each alternative execution scenario was associated with a set of features (differences between two execution scenarios) and their respective execution cost. The impact of these features on business process execution cost is analysed by performing a regression analysis, where a low regression estimate value indicates a high impact on execution cost reduction. Finally, insights such as: 1) features with the highest impact on execution cost; 2) the preferred task execution for a resource; and 3) the preferred resource allocation for a task can be derived from further analysing the regression result. This provides process analysts with a better understanding of the changes (to make) that may result in cost minimisation.
Experimental evaluation conducted to determine the feasibility of the proposed approaches in practice.
- Case Study: