Enter the current hour on your clock, 0 to 23
Research earns its keep when it is tested in use. These are four applied prototypes and models, built around the people inside complex projects: their safety, their wellbeing, their learning and their work.
A quiet first step, for the people on site.
A conversational prototype exploring a quieter first step toward psychosocial support for construction workers. Its safety-first interaction design was developed for an industry where reaching out can be difficult.
Feedback that points forward, at scale.
A demonstration agent for feedforward: guidance aimed at the next attempt, not just a grade on the last one. The research asks how personalised assessment guidance can remain feasible across large cohorts.
Site safety know-how, on demand.
A chatbot developed to share construction safety knowledge and make guidance, regulations and site experience easier to retrieve in plain language.
Cost & schedule certainty for sustainable builds.
A published machine-learning model for predicting final cost and duration from early green-building project data, trained on 198 completed Hong Kong projects and implemented in industry by Yau Lee Construction Company Limited.
One stack carries a project from a question to a testable form: modelling and text mining in R and Python, agents orchestrated and observed, then evaluated through demonstrations, publications or applied use.
