Optimising the layout of the equipment and connecting pipes that form a chemical processing plant is an important and complex problem for the oil and gas industry. The aim of this optimisation problem is to minimise the total cost of the plant while ensuring its safety and correct operation. The complexity of the problem is such that it is still solved manually, taking multiple engineers years to complete and yielding solutions of unknown quality.
Monash is collaborating with Woodside to develop state of the art optimisation and visualisation technology, that is already capable of producing 3D (near) optimal plant designs in just a few hours for medium and large size plants. I will discuss the different components and technology developed during our collaboration, together with their advantages and current limitations. I will also discuss how obtaining these designs has opened up many research questions regarding solution diversity, explainability, and interactive optimisation.
Faculty of Information Technology, Monash University
Maria Garcia de la Banda is a Professor at the Faculty of Information Technology, at Monash University and has more than 25 years of experience as an academic. Prof. Garcia de la Banda’s PhD, awarded in 1994, won the Universidad Politécnica de Madrid’s Best PhD Award. In 1997 she was awarded the first and only prestigious Logan Fellowship for the Faculty of Information Technology at Monash University, a role she held for 6 years. In 2005 she won, with Peter Stuckey, the International Constraint Modelling Challenge. She has been an elected member of the Executive Committee of the Association of Logic Programming, and of the Association for Constraint Programming, of which she is currently president. From 2013 to 2016, she was the Deputy Dean of the Faculty of Information Technology at Monash University, and prior to this she was the Head of its Caulfield School of Information Technology.
Prof. Garcia de la Banda has extensive research experience in the modelling and solving of combinatorial problems, with particular emphasis on the (semi-)automatic analysis and transformation of constraint programming models, and has been involved in different real-world problems, including the optimization of the layout of LNG plants, and the optimization of Melbourne’s water flow system.