SCALE@NTU

Research Focus

Research and development efforts in SCALE@NTU focus on AI, data analytics, robotics and smart computing. The lab will allow researchers to develop applications for use in the areas of public safety, smart urban solutions, transportation, healthcare, and manufacturing. It also aims to resolve various challenges faced by cities in keeping their infrastructure facilities in tip-top condition.

Specifically, this lab will create novel technologies and innovations under the following three research themes:

1) Anticipatory Analytics and Services; 2) Edge Intelligence, and 3) Condition-based Maintenance. 

These are essential capability development areas for Singtel/NCS to build up key differentiation solutions for Smart & Safe City. They are also ongoing strategic research areas of NTU being a leading institution in Cognitive Systems & Artificial Intelligence.

Research Theme 1 – Customer: Anticipatory Analytics and Services

This research theme focuses on “Customer” dimension of Smart & Safe City. R&D will be performed to create technologies that enable enterprises to go beyond predicting and to make real-time anticipation. This requires significant increase in certainty through research in data intelligence, automatic understanding of customers' behaviors and intentions and real-time resource management. Traditional service providers or businesses are people-centric in which manpower is used to answer customer’s request, analyze their interests, and provide suggestion to the customers. With the dramatic advances in A.I. technologies, it is more effective to move the people-centric work mode to seamless collaboration between A.I. and people such that intelligent services can be more reactive and personalized to customer requests.

Programme Director: Prof Ong Yew Soon

Research Theme 2 – Environment: Edge Intelligence

This research theme focuses on “Environment” dimension where increasing variety of sensors are being deployed. Situational management scenario needs analytics at the edge of the sensor. Edge analytics methods are proposed to achieve basic data processing and analytics, such as facial recognition. Although edge analytics technologies can be used to “aware” situational information (e.g., who are involved in an event), it is more essential to understand the situation (e.g., what is happening, what is the reason of an incident?) for making actionable suggestions in situational management. Furthermore, multiple types of sensor data (e.g., video, audio, etc.) make the edge analytics more complex.

Programme Director: Prof Gan Woon Seng

Research Theme 3 – Machine: Condition-Based Maintenance (CBMx)

This research theme focuses on the “Machine” dimension. Future Smart Cities must address sustainability, given the explosive scale of IoT and sensors. Continuing traditional periodic maintenance without considering the actual condition, taxes the resources as systems become more complex. Joint Lab R&D aims to create innovative technologies that allow effective prediction of system condition to enable optimal system function with the lowest cost of maintenance. We plan to study A.I. solutions for condition-based maintenance. Specifically, the solution focuses on the following two areas: a) Equipment Status Analytics: Detect faults and predict the future failure/outrages of the equipment for conducting preventive maintenance by developing deep learning based techniques. b) Planning and Dynamic Resource Allocation: Automatically suggest maintenance plan (replace, repair, etc.) for components and allocate resources for the maintenance actions.

Programme Director: Assoc Prof Yeo Chai Kiat