EURO 2025 Leeds
Abstract Submission

1213. Optimizing Workforce Allocation with Evolutionary Artificial Intelligence: A Case Study in Intelligent Job Scheduling

Invited abstract in session WD-34: Applications of Knowledge Work Technology, stream Advancements of OR-analytics in statistics, machine learning and data science.

Wednesday, 14:30-16:00
Room: Michael Sadler LG10

Authors (first author is the speaker)

1. Abimbola Falodu
Computer Science and Digital Technology, Aston University
2. Alina Patelli
Computer Science and Digital Technology, Aston University
3. Aniko Ekart
Computer Science, Aston University
4. Charlotte Burton
Business Development and Marketing, Thames Laboratories

Abstract

On-site maintenance job scheduling presents a complex optimisation challenge which, can cause suboptimal resource utilisation, workforce dissatisfaction, and increased operational costs. Traditional scheduling methods, widely employed across the engineering services industry, are resource heavy, and struggle to manage real-time constraints, delaying decision-making and prolonging service lags.

We introduce a novel evolutionary framework to make workforce allocation effective and efficient under realistic operational constraints. The framework comprises an innovative genetic algorithm with bespoke schedule encoding mechanisms, a clustering heuristic that improves solution quality via efficient initialisation, and a computational strategy that leverages parallelised distance matrix calculations. Dynamic workload balancing, priority-based job allocation, and a solution caching and retrieval mechanism across generations enable real-time adjustments when disruptions occur and increase the framework’s operational feasibility.

Engineering service provider Thames Laboratories has automated their legacy scheduling system by adopting our framework, thus achieving significant improvements in organisational KPIs: reduced travel times and fuel consumption, balanced workloads, and streamlined callout attendance. This successful case study offers concrete proof that the systematic commercial adoption of robust AI tech leads to significantly improved operational efficiency.

Keywords

Status: accepted

Back to the list of papers