Project Summary
FYLD is an AI-powered work execution platform targeting the UK rail maintenance sector’s £150 million addressable market for preventable delays. While FYLD’s core platform already delivers proven results across infrastructure sectors, this project proposes developing a rail-specific predictive AI model to forecast late handbacks and job blockers in real-time. The phased approach begins with pilot deployments (2025-2026) partnering with Tier 1 contractors to collect operational data and validate the model, followed by sector-wide scaling (2028-2029). Conservative projections show 1% efficiency gains could save Network Rail £29.5 million annually, with revenue reaching £11 million by Year 5, positioning the UK as a global leader in predictive rail maintenance innovation.
Project Achievements
FYLD has established a live, operational platform currently delivering measurable improvements across infrastructure sectors including highways and utilities. The core system features GPS-enabled Command Centre capabilities, digital RAMS, video-based risk assessments, job blocker tracking, and offline functionality with 24/7 technical support. A significant achievement includes developing an OFGEM-funded “Predictive Safety Interventions” AI model, demonstrating proven capabilities in predictive analytics. One customer captured over 16,000 hours of job delays, using FYLD’s insights to redesign processes and reduce completion times. The platform offers seamless integration through open APIs and operates with ISO 27001/SOC 2 compliance. FYLD has validated its work execution technology in regulated industries, establishing foundations for rail sector deployment. Initial discussions with a major Welsh network infrastructure maintainer have confirmed shared interest in piloting the predictive delay model, positioning FYLD for real-world rail environment testing.
Conclusions
FYLD addresses a conservatively estimated £150 million addressable UK rail maintenance market, representing 20-25% of total disruption-related costs where predictive tools can influence outcomes. Even modest improvements deliver substantial value: 1% efficiency gains could save Network Rail £29.5 million annually across Schedule 4/8 payments and maintenance spend, while 10% delay reductions generate £6.6 million economic benefits for employers and £18 million societal value for passengers. No direct competitor offers FYLD’s combination of real-time job tracking, AI-driven delay prediction, and frontline enablement. The solution aligns strategically with Great British Railways vision, DfT’s Digital Railway strategy, and government rail reform priorities. Beyond UK deployment, global rail maintenance markets valued at $95 billion by 2030 present significant export opportunities, particularly in India, Europe, and Scandinavia where human-led maintenance faces similar efficiency challenges, positioning the UK as a global innovation leader.
Next Steps
Success would be made possible with a trial. The initial 12-18 months focus on securing pilot partners in high-delay-exposure regions, prioritizing the engaged Welsh network maintainer and Tier 1 contractors including Amey, VolkerRail, and Colas Rail. Months 1-3 involve finalizing pilot agreements and engaging DfT/Innovate UK for innovation funding. Months 3-6 deploy FYLD core platform to live field teams, capturing operational data on delays, handbacks, and job blockers while initiating AI model training. Months 6-12 validate pilot performance, measure baseline impacts, and iterate based on feedback to produce ROI case studies. Beyond 18 months, structured commercial offerings target regional Network Rail frameworks and national contractor portfolios. Critical enablers include accessing historical delay data, recruiting rail domain specialists, and engaging RSSB/ORR for regulatory alignment. Investment requirements total £750k-£1.2m for AI development, data infrastructure, delivery capacity, and market readiness, with breakeven projected by Year 4-5.
