At Frontline, we have always believed that good data is only useful if it can be understood and applied. As a software company focused on unlocking the value of operational data, we build tools that help organisations extract meaningful insights from complex, often messy information. When we joined the BEAMP programme, we were looking for a real-world challenge that could push us to refine our technology - and we found that in a collaboration with Penta-Ocean.
Penta-Ocean Construction is a major player in the global construction industry, known for large-scale marine and infrastructure projects across Asia. Their planning teams had identified a persistent bottleneck: too much time spent manually analysing old project schedules. The data was there, spread across files and formats, but retrieving useful insights required hours of repetitive, error-prone work. It was clear that with the right tools, this process could be transformed.
Working together, we co-developed a web-based platform tailored to the needs of Penta-Ocean. The aim was to make schedule analysis faster, smarter, and more intuitive. What made our solution stand out was the integration of AI-powered vector search, built using language models of OpenAI. Instead of manually scrolling through spreadsheets, users could now search for specific activities and instantly find similar ones across historical schedules.
The experience is simple. Project planners upload past project files to the platform and type in an activity they want to review. The system identifies and groups similar activities, drawing from diverse sources and terminology. Results can be refined using filters, and the platform automatically generates visual performance statistics, turning hours of manual effort into just a few minutes of work.
The tool has already proven especially valuable in the tender preparation stage, where time is limited and access to past benchmarks can make a meaningful difference. It also supports better tracking of actual performance against planned schedules, helping teams identify gaps and improve forecasting.
For us, BEAMP provided much more than a pilot. It offered structure, and the rare opportunity to co-create a solution alongside a leading construction firm. We were not just testing a product but shaping it with real user feedback in a live project environment. That kind of collaboration helped us build something grounded, practical, and immediately useful.
Looking ahead, we see strong potential to adapt this platform for other firms in construction and infrastructure. Each collaboration brings new insights, and we are committed to evolving the solution into a widely applicable, industry-grade tool that helps organisations turn their own data into better decisions.
Ruiqi Ding
CTO, Frontline
