Big data-based ""AI inspector"" for gauging inert contents at the off-site construction waste sorting facilities in Hong Kong

Professor Lu, Weisheng Wilson   (Principal Investigator (PI))

Miss Lee Mun Wai Wendy   (Collaborator)

Professor Xue Fan   (Co-Investigator)

24

2020-09-01

2022-08-31

485000

Big data-based ""AI inspector"" for gauging inert contents at the off-site construction waste sorting facilities in Hong Kong

""AI inspector"", Big data-based, Hong Kong, inert contents, off-site construction waste, sorting facilities

Building and Construction

ECF Project 111/2019

Environment and Conservation Fund

2019

Completed

Offsite sorting facilities (OSFs) play a key role in the Government’s strategic initiatives to strengthen construction waste management in Hong Kong. By paying a levy of HK$175/ton, contractors can deliver their construction waste to OSFs for sorting if the waste contains more than 50% of inert contents by weight. As a yardstick of successful operation of the OSFs, a clever inspection methodology has been developed to ensure that the waste truly meets the criterion, i.e. >50% inert contents. However, the Auditor’s report in 2016 pronounced the ineffectiveness of the methodology, as the inert contents sorted had always been below the bar. The Government has strengthened the inspection methodology several times, but its effectiveness had ""diminished"" quickly. It seems that the methodology is oversimplified and particularly vulnerable under the relentless cat-and-mouse game between the regulators and waste haulers. This proposed project aims to develop a more effective ""AI Inspector"" based on big data analytics (e.g., fuzzy set theory and Bayesian probability model) to gauge the inert contents acceptable at OSFs. It is expected that the ""AI inspector"" will fix the loopholes in existing methodology without necessarily installing any new equipment. It will significantly help enhance construction waste management in Hong Kong.