Scholarlink

Manual job scheduling is a resource-consuming operation that routinely results in sub-optimal customer service, inefficiently utilised, dissatisfied workforce, and decision-making that is slow to adapt to real-time developments; all at a high environmental cost. These symptoms have long been affecting the engineering service industry, with on-site maintenance providers frequently reporting operational losses due to the cumbersome logistics of dispatching engineers to attend calls at client locations. Intelligent scheduling via evolutionary Artificial Intelligence offers a promising route towards finding effective solutions to this host of problems. We have leveraged this potential to build an innovative end-to-end scheduling optimisation algorithm featuring customised components: (1) a tailored Evolutionary Algorithm equipped with a robust schedule encoding and decoding mechanism as well as bespoke genetic and fitness-tuning operators; (2) a clustering heuristic that groups jobs according to their proximity to relevant engineers leading to an initial population of higher quality candidate solutions, (3) a computational efficiency booster performing calculations and caching of distance and duration matrices by means of parallel computing, and (4) a persistent solution storage and retrieval mechanism optimising real-time alternative timeslot proposal. These have been carefully configured, enhanced, and combined to produce a robust software platform that evolves job schedules meeting realistic, dynamic operational restrictions. We demonstrate, based on evidence produced by running a comprehensive set of experiments, that our algorithm efficiently improves the key performance indicators of a real-world on-site engineering services provider, Thames Laboratories. We are successfully automating the company’s customer attendance logistics, enabling better decision-making support for the organisation’s leadership. This case study highlights the significance of our AI-powered algorithm’s impact on transforming the relevant industry for the better, in full alignment with environmental sustainability best practice, and provides a strong argument for larger scale distribution and adoption across commercial and public administration sectors alike.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Intelligent Job Scheduling: Better Service for Less Carbon

  • 章节

  • 发表日期：

    2025年

查看全文
  • Alina Patelli,
  • Abimbola Falodu,
  • Anikó Ekárt,
  • Charlotte Burton

摘要

Manual job scheduling is a resource-consuming operation that routinely results in sub-optimal customer service, inefficiently utilised, dissatisfied workforce, and decision-making that is slow to adapt to real-time developments; all at a high environmental cost. These symptoms have long been affecting the engineering service industry, with on-site maintenance providers frequently reporting operational losses due to the cumbersome logistics of dispatching engineers to attend calls at client locations. Intelligent scheduling via evolutionary Artificial Intelligence offers a promising route towards finding effective solutions to this host of problems. We have leveraged this potential to build an innovative end-to-end scheduling optimisation algorithm featuring customised components: (1) a tailored Evolutionary Algorithm equipped with a robust schedule encoding and decoding mechanism as well as bespoke genetic and fitness-tuning operators; (2) a clustering heuristic that groups jobs according to their proximity to relevant engineers leading to an initial population of higher quality candidate solutions, (3) a computational efficiency booster performing calculations and caching of distance and duration matrices by means of parallel computing, and (4) a persistent solution storage and retrieval mechanism optimising real-time alternative timeslot proposal. These have been carefully configured, enhanced, and combined to produce a robust software platform that evolves job schedules meeting realistic, dynamic operational restrictions. We demonstrate, based on evidence produced by running a comprehensive set of experiments, that our algorithm efficiently improves the key performance indicators of a real-world on-site engineering services provider, Thames Laboratories. We are successfully automating the company’s customer attendance logistics, enabling better decision-making support for the organisation’s leadership. This case study highlights the significance of our AI-powered algorithm’s impact on transforming the relevant industry for the better, in full alignment with environmental sustainability best practice, and provides a strong argument for larger scale distribution and adoption across commercial and public administration sectors alike.