Alchera Technologies has secured £2.7 million in seed funding to expand its artificial intelligence software for transport infrastructure planning and network analysis. The UK company develops technology that brings together road, vehicle and public transport information, allowing infrastructure operators to examine network performance and assess the effects of planned works or changes to traffic arrangements.
The funding comprises £2 million in investment led by Maven Capital Partners through the British Business Bank’s South West Investment Fund, with participation from the Bayes Entrepreneurship Fund and The FSE Group. Innovate UK’s Investor Partnerships programme has provided a further £700,000 in matched funding, while existing investor Cambridge Applied Research also participated.
Alchera intends to accelerate product development, improve how its platform can be introduced into customer systems and expand adoption among highways authorities, transport operators and engineering advisers. The company says its software can reduce analysis work that previously took weeks to decisions made within the same day, although this remains a company performance claim rather than an independently measured result across all applications.
Transport networks generate data through numerous systems. Traffic sensors, connected vehicles, bus location services and traffic management equipment provide information about movement and operating conditions, while weather and other contextual records can help explain variations in network performance.
These sources are frequently maintained by different organisations or stored in systems that were developed for separate purposes. A traffic authority may have information about junction operation, while a bus operator retains detailed vehicle movement records and a maintenance contractor holds the timetable for planned roadworks.
Bringing those records together requires information to be matched across locations, time periods and network features. A measurement collected at one junction may not describe conditions along an entire route, while differences in how data is recorded can complicate comparisons between sources.
Alchera’s platform is designed to integrate those datasets and apply proprietary machine learning methods to identify traffic patterns and support decisions. Its applications include roadworks scheduling, bus network analysis and planning for disruption caused by changes to infrastructure or operating conditions.
Road maintenance and construction can alter a network’s available capacity. A lane closure may reduce the flow of vehicles through a section, while temporary traffic signals can change the time required to pass an individual junction. The consequences depend on the location, traffic demand and alternative routes available.
Planners need to assess those effects before work begins, particularly where several projects could affect neighbouring roads. Two restrictions that can be accommodated separately may create greater disruption when implemented together because traffic diverted from one route encounters reduced capacity elsewhere.
Analysis can help identify those interactions by examining how vehicles currently move through the affected areas. Historical journey times, traffic counts and other records provide evidence of normal operating conditions, while information about planned interventions can be used to test different scheduling arrangements.
The reliability of the results depends on the representativeness of the information. Traffic patterns change throughout the day and between weekdays, weekends and seasonal periods. A model developed using a limited set of observations may not capture the conditions encountered during an unusual event or a major change in travel behaviour.
Data quality introduces further constraints. Missing observations, inconsistent timestamps and changes to road layouts can alter the meaning of a recorded measurement. The platform therefore needs dependable information about the location and circumstances of each dataset before combining it with other sources.
Machine learning can identify relationships within complex information, but it cannot eliminate uncertainty about future traffic conditions. Its predictions are influenced by the available training data and the assumptions used in the model, making comparison with actual operating results important when a system is deployed.
Transport infrastructure decisions also involve considerations beyond vehicle movement. Maintenance activities may require safe working space, access for construction equipment and compliance with restrictions governing the road. A predicted reduction in congestion cannot override those engineering and safety requirements.
Bus operations provide another application for the technology. Information about vehicle location and journey times can identify recurring delays, unreliable sections of a route and variations between expected and actual operating conditions.
Combining that information with traffic records may help authorities distinguish between congestion affecting all vehicles and problems associated with a particular service or route configuration. The outcome can support assessments of signal timing, road layouts or the scheduling of maintenance activity.
However, an observed delay does not establish its cause without further investigation. Passenger boarding, service frequency, road geometry and temporary incidents can all influence bus performance, so the underlying circumstances must be considered before changes are recommended.
Alchera is also developing applications involving longer-term transport planning and network resilience. Infrastructure operators need to consider how roads and public transport may perform when demand changes, construction activities overlap or weather events restrict parts of a network.
Climate-related disruption introduces particular uncertainty because flooding, high temperatures and other conditions can affect different assets in different ways. Historical information can support the assessment of vulnerabilities, but events outside the range represented by previous data may require separate engineering analysis.
The platform provides analytical support rather than replacing physical asset inspection or structural assessment. A traffic model can help estimate the disruption caused by closing a bridge, but it cannot determine the bridge’s structural condition or establish whether repair work is technically necessary.
Alchera’s approach is intended to reduce the effort required to prepare usable information for those decisions. The company says transport teams can move from manually combining datasets over extended periods to evaluating relevant evidence within a shorter operating window.
Chief executive and co-founder Anna Jordan has identified fragmented information as a significant obstacle for authorities responsible for roads and bus networks. The company is seeking to make data from different systems more accessible to the people planning infrastructure work and responding to network problems.
External access and integration are important to that model. Customers need ways to connect their existing data sources without replacing every system already in service, while maintaining appropriate arrangements for information quality, permissions and continuing updates.
As more information is added, software must also distinguish between changes in the transport network and inconsistencies caused by how data is collected. A new sensor or revised reporting method may alter recorded patterns without representing a corresponding change in actual traffic conditions.
Investment in the platform will support development and wider deployment, but Alchera has not announced a new physical transport infrastructure project, a construction contract or a quantified reduction in congestion arising from the financing.
The £2.7 million round provides additional resources for software intended to support decisions about existing roads and public transport systems. Its value in operation will depend on how accurately the platform combines changing data sources, how reliably its analysis reflects actual network behaviour and whether infrastructure teams can use the results alongside established engineering and safety requirements.
