Where: 080.03.015

When: 11:30 to 12:30, Friday, 08 June 2018.

Abstract: Open pit block scheduling is a challenging optimisation problem in the mining industry. The aim is to create the best possible open-cut mine plan in order to maximise the net present value of an ore body. This talk discusses a novel mixed integer programming based approach - Merge search - to tackle this problem. The results show that Merge search is able to provide better quality solutions than those available in the literature and a parallel branch & bound search improves upon the best known upper (relaxed) bounds known to date. Furthermore, a parallel implementation of Merge search is able to obtain good solutions more rapidly.

Bio: Since completing his PhD in combinatorial optimisation (2011), Dhananjay has been working at the interface between academia and the industry. He has worked problems originating in the mining industry, scheduling and biosecurity. In order to solve these problems, he has developed and investigated methods derived from artificial intelligence (constraint programming, nature-based meta-heuristics, Bayesian networks) and operations research (mathematical programming). He currently works at MAXIMA (applied math institute at Monash University) and explores ways in which industrial problems can be solved efficiently.