Graph-based Process Discovery containing Invisible Non-Prime Task in Procurement of Animal-Based Ingredient of Halal Restaurants

Abstract

A process model based on animal-based ingredients' running procurement will help halal level examiners check halal implementation based on Halal Critical Control Points (HCCP). Process discovery is a study for automatically forming a process model based on a log of running processes. There are several algorithms of process discovery, such as Alpha++ and Alpha#. However, the procurement of animal-based ingredient processes has an invisible non-prime task that has not been discussed in the existing algorithms. This research proposes a graph-based process discovery algorithm to form a process model containing invisible non-prime tasks based on the procurement processes. The log of procurement processes is obtained by using a business process management application, i.e., ProcessMaker. This research evaluates the proposed graph-based algorithm by comparing it with Alpha++ and Alpha# based on fitness and precision. The evaluation verifies that the proposed graph-based algorithm has a better quality of the obtained process model than Alpha++ and Alpha#. The fitness and precision of the graph-based algorithm are 1 and 1. On the other hand, the precisions of Alpha++ and Alpha# are 0.43 and 0.43, respectively.