Application of AI to Improve Workforce Allocation and Efficiency in British Ports

The shipping industry is responsible for around 940mt of CO2 emissions annually, at least 2.5% of the world's total CO2 emissions (UKRI, 2021) and 90% of everything we consume is moved by sea. However, the shipping industry remains a laggard in terms of digitalisation and the development of disruptive, data-driven, real-time analytics to improve and streamline operations.

Labour is a vital component within port terminals and accounts for over half of a port terminal's variable costs. Port operations are undermined when there is a shortage of workers, leading to substantial delays in vessel loading and discharging operations, subsequently limiting the further supply chain. There are major challenges in labour allocation and long-term planning. Early forecasting on port operations using data and statistics will give job allocation teams the information required to ensure that they are aware of future skill requirements and are able to maximise efficient resource utilisation.

UK transport and logistics sectors are not digitally enabled and expected to be short by 400,000 employees in 2026\. Labour shortages are projected to be a limiting factor in the productivity of UK ports in the years to come, despite the jobs being labelled as well paid and 50% more productive than other sectors (Maritime UK 2020). 59% of employees have said that they do not think maritime pays enough, nor was it technologically/digitally advanced enough to attract people into the sector.

The shortage of workers reduces operational capacity and efficiency resulting in delays in goods at ports. This exacerbates supply chain impacts which are felt far beyond the port. A major factor for this shortage is a lack of statistically driven labour forecasting tools that can provide job allocations based on available skills and contractual constraints, training and hiring advice using data from scenario-based modelling. Currently, labour forecast modelling lacks granularity, is slow and is undertaken biannually at most UK ports; leading to a reactive rather than proactive approach to dealing with changes in demand.

This project will utilise advances in AI techniques to dramatically reduce the time it takes to produce labour scheduling and forecasting models using contractual data within Ensemble Analytics' Athena platform. Currently, this is carried out manually and is very time consuming. It will increase the functionality of our existing tools and do so in a scalable manner, so that they can be rolled out to ports and ultimately to other wider blue-collar industries.