ABSTRACT
In an organisation, effective business process management (BPM) may aid in achieving organisational objectives. In order to improve organisational effectiveness and boost overall performance and competitiveness, it is crucial to properly manage these processes’ lifetime. One of the most significant issues in BPM may be resource allocation (RA). Task similarity has received little attention despite this topic being the focus of much research. This study suggests a unique approach to resource allocation and digital job optimisation in automated BPM. Here, the Markov decision entropy Q-cluster Bayesian network allocates business process resources. Then, a heuristic swarm colony vector optimisation model optimises the network. Experimental investigation is conducted regarding operating cost, QoS, scalability, resource utilisation, and validation accuracy. We suggest addressing these issues by applying convergence speedups consistently throughout the learning phases and suitable initialisation for the early stages, and we report our initial experimental findings.
