Local authorities need to support their funding requests with high-quality data. The problem is that they can't obtain this data at the required speed, consistency, and quality to achieve this.
The UK’s critical infrastructure includes approximately 246,500 miles of roads and 80,904 bridges. Each of these individual assets requires monitoring, maintenance, and management to keep them operational, and for the vast majority, this falls under the remit of one of the 317 local authorities. However, when we look at the deteriorating condition of this critical infrastructure, it’s clear that these local authorities are fighting a losing battle when it comes to effective infrastructure management. The solution to this problem is far from straightforward, but a good place to start is at the inspection level, where artificial intelligence can play a key role in turning the tide.
Our critical infrastructure is approaching a crisis point. A vast number of roads, tunnels, bridges, and other structures have reached an age where deterioration accelerates. The RAC Foundation highlights that 2,928 of the nation's road bridges, 4% of the total, are “substandard,” meaning they cannot carry the heaviest vehicles, which impacts freight, agriculture, and overall network efficiency.
Meanwhile, the 2025 ALARM survey reports that less than half (48%) of the local road network is reported to be in good structural condition, with the remaining 52% stated to have less than 15 years’ structural life remaining. 17% of this road network is reported to be in “poor condition”.
It’s estimated that addressing the backlog of substandard bridges would cost £6.8 billion. For the road network, this number is £16.81 billion to bring the road network up to its ‘ideal’ conditions, work that would take 12 years to complete. This backlog is only getting worse, having increased by 42% from £11.8 billion since 2016.
Local authorities are acutely aware of this problem, but their ability to remedy it is being hampered on many fronts. In the last year, highway maintenance budgets saw a real-terms cut of 4.1% in England and Wales, to an average of £26 million per authority. Authorities can apply for funding increases, but these need to be backed by structural health data that can be clearly articulated to stakeholders, and at present, several factors hamper these efforts.
There is a wealth of historical information on structures collected over decades of inspections. However, this data takes numerous forms, ranging from hand-written notes and printed files to digital reports featuring high-res imagery and video. Finding this information can be a challenge in itself, let alone interrogating it to gain an understanding of an asset’s history. Furthermore, this lack of consistency makes it difficult to zoom out and compare assets at the portfolio level and identify which ones require intervention.
Ultimately, this creates a fragmented understanding of asset health at both individual and portfolio levels. For local authorities to articulate the need for increased funding, they need the data that will show that their assets are deteriorating to the point of needing maintenance, and that the optimal moment to make this intervention has arrived. The better the understanding, the more compelling the case for funding can be, but if local authorities don’t understand which assets need attention, then they certainly can’t communicate this to funding decision-makers.
One way to address this problem is to focus on data capture deficiencies at the inspection level. If we can ensure that data is captured in a comprehensive, consistent, and high-quality way, it will not only improve our understanding of structural health but also lay the groundwork for AI-powered condition intelligence.
Mind Foundry Windward is a condition intelligence platform. At the heart of the platform is a model for quantifying the nature and rates of asset deterioration over time. The resulting asset condition information powers explainable Machine Learning models, which help prevent failures, reduce unnecessary expenditures, and keep systems operating.
But to harness this condition data, it must first be collected. Mind Foundry Windward is designed to modernise in-person inspections without disrupting established workflows. It puts AI in inspectors' hands, enabling them to capture accurate information about defects on a structure and their contribution to the asset’s overall health by capturing high-quality images, tagging defects, locating issues, and logging condition notes.
In addition to mobile, it features a desktop application where images, notes, and defect changes can be combined into data-rich reports that are easy to analyse and interrogate. Having high-quality, consistent data readily accessible gives a comprehensive overall picture of an asset's current condition, as well as deterioration over time. Thus, what begins as a better understanding of one asset’s health gained from an inspection can become a drastically enhanced understanding of how an entire portfolio of assets is changing.
This understanding can provide the foundation for a collaborative effort to address the condition crisis. The infrastructure sector is vast, and the problem is complex; therefore, no single entity will have all the necessary expertise, equipment, and resources to solve it alone. This is where the importance of partnerships comes into play, encompassing local authorities, central government, infrastructure consultancies, and SMEs. With effective collaboration, we can build solutions that are greater than the sum of their parts, but to solve the problem, we must first understand it.
To address the crisis of deteriorating critical infrastructure, local authorities must present compelling, data-driven cases to secure the necessary funding. Mind Foundry Windward establishes the foundation of data that these authorities require to comprehend their portfolios and effectively communicate this understanding to their stakeholders. In doing so, it alleviates some of the primary difficulties with modern asset management.
Currently, asset management is based on a straightforward ranking system. Each asset is given an overall condition score based on the severity and extent of all the defects on that asset. From there, assets are lined up from “worst” to “best”. Asset engineers then work down the list, starting with the asset that performs the worst overall. They identify the worst defect on this asset and initiate maintenance accordingly.
This system creates two main problems:
This is where communication is key. Asset engineers must explain these risks to decision-makers, funders, and stakeholders in a manner that resonates with each of them. This means making the risks visible, showing the consequences of inaction, and enabling better communication between the people who see the defects and the people who control the budgets.
By capturing data with the quality and consistency required to assess defects accurately, local authorities can begin to assemble a comprehensive picture of a structure’s health. Then by building a picture of how a defect has progressed and, critically, how it might develop if left unchecked, they can prioritise maintenance accordingly, allocate resources more effectively, and ultimately become more efficient in their maintenance operations.
Armed with this information, local authorities can be empowered to make compelling, evidence-based appeals for funding at a time when economic conditions are straining budgets and pressure to make informed, accurate decisions is at an all-time high. Against a backdrop of deteriorating critical infrastructure, having the tools to extract every ounce of insight possible from structural condition data is the first step to regaining control.
Click here to find out more about Mind Foundry Windward.