IBM Labs Uses Complex Math To Fight Natural Disasters

IBM Labs on Tuesday introduced a software system designed to help solve the problems of natural disasters by using some of the same complex algorithms IBM uses in managing its corporate customers.

The high-level math problem known as IBM's "stochastic optimization model," uses math equations to help speed up and simplify complex tasks into everyday life -- such as determining the fastest route to deliver packages, detecting fraud in health insurance claims, and scheduling supply chain and production at a manufacturing plant.

The same methods could be used, researchers said, to efficiently deliver a large number of critical resources to a range of disaster event scenarios such as wildfires, floods, and disease.

"Most real-world problems involve uncertainty, and this has been the inspiration for us to tackle challenges in natural disaster management," Dr. Gyana Parija, a senior researcher and optimization expert at IBM India Research Laboratory, New Delhi, said in a statement.

In the case of flooding, for example, IBM said its stochastic programming model would number crunch various flood scenarios, resource supply capabilities at different dispatch locations, and variable costs of various flood-management resources to predict outcomes and help governments and relief workers better manage various risk measures.

The underlying optimization models and algorithms were initially prototyped on a large U.S. government program. The current version includes simulation models to assess and evaluate the impact of several strategies.

The software was developed by a team of researchers with IBM Research Labs in New York and India working with business experts from IBM's Global Business Services.

IBM said the goal now is to equip emergency responders of various sizes with tools for better strategic planning.