Research Interest |
||||||||||||
My research interests are mainly focused on
employing latest search and machine learning techniques (Meta-heuristics,
Hyper-heuristics, Evolutionary Algorithms, and Machine Learning) as well as
conventional optimisation approaches (LP/IP/MIP, Branch & Bound, Cutting Plane,
Nonlinear Programming) to solve various real-world problem applications,
including transportation scheduling, space planning, and timetabling. I am
particularly interested in studying how search algorithms can
automatically adapt to different environments and circumstances by
learning from either during or after search processes. My research
areas include:
|
||||||||||||
|
||||||||||||
Current Project: I am currently working on a number of projects, mainly focusing on developing models and algorithms to optimise current transportation logistics systems and urban transportation systems, for example the smart taxi project Previous Projects:PLATFORM: Towards More General Optimisation/Search System Funding body: EPSRC (Ref: GR/S70197/01) Period I involved: 10/2005 - 08/2007 This research focuses upon investigating and developing a fundamentally more general and adaptive search/optimisation methodology (hyper-heuristics) than exists today. This goal represents one of the key research strategies that are currently facing the international optimisation community. Outcome of this research would be adaptive algorithms which learn to adapt to different problems and situations by various learning techniques. The results of the research would have significant impact on decision support system development and its deployment across a wide range of diverse applications and industries. Specifically, I investigated and developed a simulated annealing hyper-heuristic framework that can be readily adapted to various problem solving scenarios. Various problems have been tested, including university course timetabling, nurse rostering, bin packing an. Later, I worked closely with CRT (Centre for Research on Transportation) in Universite de Montreal in Canada, investigating the role of hyper-heuristics in the development of next generation transportation scheudling and planning systems. CRT is a world leading research group in transportation systems, network optimisation, routing and scheduling. Investigation of Novel Methods for Optimising Retail Shelf Space Allocation (PhD Project) Funding body: EPSRC (Ref: GR/R60577/01) Period: 08/2002 - 09/2005 The retailing industry is an extremely competitive arena. In this project, we investigated modelling and latest optimisation techniques (including heuristic, meta-heuristics, hyper-heuristics, LP relaxation, generalised reduced gradient algorithm, decomposition approaches, etc.) in order to improve inventory control and shelf space management in a retail outlet. The research results of the project would help to develop next generation planograms software which could automate and optimise some of the retail space planning and operation decisions. The project is in collaboration with Tesco (the largest retailer in the UK), Retail Vision and SpaceIT Limited. Project assessment: Tending to outstanding. Nesting of Two-dimensional Irregular Shapes Utilising The No-Fit Polygon Period: 09/1999 – 04/2002 The research consists of the main body of my master thesis. The nesting algorithm used the no-fit polygon techniques (obtained by calculating the Minkowski sum) and a genetic algorithm to optimise the packing. Network Based Educational MIS System (NBE-MIS) Period: 09/1999-08/2000 This software was developed on behalf of the Northwestern Polytechnical University where I obtained my bachelor degree and master degree. The software provides the convenient interface which allows the university administrative stuff to plan and manage the majority of the teaching activities and student information via campus networks. The developing tools are SQL/Server7.0 , Visual Basic6.0 and network technology. This project was in collaboration with three other teammates. |