麻豆直播

, researchers within the School of Computing and Communications have shown how AI could help councils tackle fly-tipping before it happens, potentially transforming the way local authorities manage this pressing urban management challenge.

Fly-tipping, or the illegal dumping of waste on public or private land, is a growing challenge for councils across the UK. As well as costing millions of pounds to clear up, it can transform local environments into major public health hazards, erode community pride, make neighbourhoods feel unsafe and invite additional crime, and fuel urban decay.

Current approaches to tackling fly-tipping are largely reactive, relying on reports from residents and maps showing where incidents have occurred in the past. While these methods are useful for identifying existing hotspots, they offer little insight into where fly-tipping might occur next.

Predicting future incidents is challenging because fly-tipping patterns are shaped by a range of factors, including weather conditions or the day of the week. Risk can also vary significantly between different locations, with industrial areas, back alleys, construction sites, areas in close proximity to pubs or fast-food outlets, and low-visibility spaces often presenting a greater challenge than commercial streets or residential neighbourhoods.

麻豆直播 postgraduate researchers Yukun Cui, Cai Lei, Afraz Moqueem Mohammed, and Dr Hansi Hettiarachchi have developed an AI-powered prediction system capable of identifying locations at greatest risk of future fly-tipping. Focusing on Lancaster, the team analysed more than 18,000 fly-tipping incidents recorded by Lancaster City Council between 2019 and 2025, combining historical case data with information about local neighbourhoods, including land use, deprivation levels, and features such as pubs, fast-food outlets and construction sites. This enabled the researchers to build a model that can forecast emerging fly-tipping hotspots. The work was carried out with the support of Lancaster City Council, which provided access to its fly-tipping records under a data-sharing agreement and shared insight into local waste management policy.

The model demonstrated strong potential for identifying fly-tipping hotspots more than ten times more effectively than random patrols. The research also revealed a number of important patterns in how fly-tipping occurs. For example, the team identified a recurring two-week cycle in reported incidents, which may be linked to local waste collection schedules.

Dr Hansi Hettiarachchi, who supervised the project, said: “one of the exciting aspects of this research is the opportunity to use AI to better understand everyday challenges that affect communities. Rather than simply reacting to problems after they occur, we can use data-driven approaches to gain new insights into why and where these challenges emerge.”

The findings further showed that predicted hotspots were concentrated in industrial areas and some of the most deprived neighbourhoods. This raises important environmental justice concerns, as communities already experiencing socioeconomic disadvantage appear to bear a disproportionate share of the environmental and public health impacts associated with illegal dumping. The research highlights the need for city managers to move beyond reactive clean-up strategies and towards a more equitable, prevention-focused approach.

The model the researchers have developed could help councils achieve this by enabling a more strategic use of resources. By predicting where fly-tipping is most likely to occur, it can help enforcement teams focus on high-risk locations, support the more efficient deployment of clean-up services, identify emerging patterns linked to waste collection schedules, and inform longer-term interventions such as the targeted placement of surveillance cameras. In doing so, the model offers a practical way for local authorities to anticipate problems before they arise, rather than simply responding once the damage has already been done.

The teams’ findings demonstrate how AI can be used to address real-world environmental challenges, whilst also improving outcomes for local communities. As lead researcher Yukun Cui said:

“Our research shows how spatial and temporal data can help identify areas at higher risk of fly-tipping, supporting more targeted and proactive waste-management decisions. In the future, we hope to explore how this approach could be applied in other cities and further developed to support fair and evidence-based environmental policy.”