IBM's research laboratories in the U.S. and India have fine-tuned technology to help model and manage natural disasters such as wildfires, floods, and diseases.
The new enhancements are to a budgeting system being developed by IBM, starting from 2003, for managing natural disaster events, with a focus on better preparedness for future uncertain disaster scenarios. The optimization models and algorithms were initially prototyped on a large U.S. government program, to deploy a large number of critical resources to a range of disaster event scenarios, said Gyana Parija, lead researcher in the Analytics and Optimization Research team at IBM India Research Laboratory in Delhi, India.
That system however only generated a single solution for each disaster scenario. The current enhancements to the budgeting system include the development of a decision support system to allow decision makers to consider multiple solutions to each disaster scenario, so that a range of alternatives can be generated by the system, IBM said Tuesday.
A model that supports multiple criteria can be used effectively in situations where there is a contention for resources, as for example when then are more than one disasters demanding resources, according to Parija. "Typically what happens in a particular disaster scenario is that you would have different budget alternatives, and at different budget alternatives, you would like to explore what kind of resource organization you can have," he added.
IBM's stochastic optimization model is designed to deal with uncertainties in data, and models with probability distributions based on historic trends, Parija said. The model can also be used to work in applications other than natural disaster management, such as asset liabilities management problems in the financial services and other business sectors, he added.