be evacuated to, and which routes should be used to evacuate them. We optimise evacuation decisions from impacted areas to shelters while considering the risk associated with different parts of the underlying road network. DISASTER RESPONSE DEPLOYMENT When you have a disaster, you have to dispatch specialists to assist the people who have been evacuated. Imagine a situation where a village or part of a village has been evacuated. You need to set-up temporary shelter facilities; you must provide medical assistance. For all of this, you need personnel – whether government employees or from NGOs – and you need to schedule them, working out the optimum way to visit the different locations they are needed. When you provide these services, not all of the specialists can arrive at the same time. For instance, before medics provide their services, somebody else has to set up the tent as a temporary clinic. So, the problem of achieving optimum scheduling of the resources available becomes more complex. Once in place, people working in disaster-hit areas generally experience two types of problem. Firstly, fatigue. The workload is huge, but emergency response workers also have lots of psychological pressure. Therefore, they need a break, and you have to take this into account when planning. The second issue is that the specialists come from different locations. You have to collect them and take them to the disaster site. That is another complex mathematical problem to be resolved. PRACTICAL APPLICATION We can build programmes to help with this, but developing the models by themselves is not enough. You can develop models and use them to produce some results, but we wanted to develop a tool that people without modelling expertise can use easily. The DSS we developed in collaboration with the University of Indonesia has a user-friendly interface. The people who need to make the decisions can enter the required information in a structured and easy to understand way. We also use information regarding the risk associated with the underlying road network. When you make the decisions to evacuate people, and to schedule disaster response personnel to reach the disaster-impacted areas, you have to consider risk. You want to schedule these movements in a way to minimise the risks to those being transported. With our system, you can visualise this information. You can see it on the map, and you can also see the trade-offs between the different objectives. The feedback we have received in Indonesia is positive. Natural disaster response will always be a complex undertaking, but through RESPOND-OR and RESPOND-OR-X we hope we can provide the tools to make it more effective, equitable and fast – helping people when they need it most. FIFTY FOUR DEGREES | 25 Konstantinos Zografos is a Distinguished Professor in the Department of Management Science, and Director of the Centre for Transport and Logistics (CENTRAL) in Lancaster University Management School. He is the Principal Investigator on the EPSCR UKRI Global Challenges Research Fund-funded RESilient Emergency Preparedness for Natural Disaster Response through OR (RESPOND-OR) project. Lancaster University members of the team were Professor Juliana Sutanto, Dr Ahmed Kheiri, Dr Aleksandr Pirogov and Dr Istenc Tarhan. Professor Zografos, Professor Sutanto (now at Monash University, Australia) and Dr Raja Fatricia participated in RESPONDOR-X. k.zografos@lancaster.ac.uk
RkJQdWJsaXNoZXIy NTI5NzM=