Figure
What the Figure Shows
Research Area 03 · Construction Decision Intelligence Lab
A construction program is a process graph that changes every shift. This area builds algorithms that certify risk on that graph, carry provenance with every decision, and spend verification effort where uncertainty is highest.
A construction program is a process graph that changes every shift: activities, interfaces, RFIs, inspections, and handoffs, with information arriving late and unevenly. Risk gets certified once, at the plan, and then drifts. Verification effort is spread evenly across everything instead of concentrated where uncertainty is highest, and digital twins pile up data that never reaches a decision.
Algorithms that certify risk on a dynamic process graph, carry provenance with every decision so a claim can be traced back to the evidence that supported it, and adapt verification effort to where uncertainty concentrates. Alongside: a maturity model for AI-enabled safety systems, digital-twin readiness diagnostics, and RFI routing simulation, so the methods have working tools behind them.
Figure
Research → Think Tank → Field Instruments
Active work
Algorithm development
A decision-making framework that models a construction program as a dynamic process graph, certifies risk at each node with a provenance trail, and concentrates verification effort where uncertainty is highest instead of spreading it evenly.
What it needs
Framework manuscript
A multi-dimensional way to assess where an organization's AI-enabled safety systems actually stand, from pilot theater to trusted operating loop.
What it needs
Still open
Research in motion
Get in touch
Co-investigators in operations research, formal methods, and machine learning; owners or general contractors willing to share process logs and schedule data from a live program.
Propose a Collaboration Offer a Site or Dataset Student Research