A continuous approximation approach for assessment routing in disaster relief

In this paper, we focus on the assessment routing problem which routes teams to different communities to assess damage and relief needs following a disaster. To address time-sensitivity, the routing problem is modeled with the objective of minimizing the sum of arrival times to beneficiaries. We propose a continuous approximation approach which uses aggregated instance data to develop routing policies and cost approximations. Numerical tests are performed that demonstrate the effectiveness of the cost approximations at predicting the true implementation costs of the policies and compare the policies against more complex solution approaches. The continuous approximation approach yields solutions which can be easily implemented; further, this approach reduces the need for detailed data and the computational requirements to solve the problem.

► We use continuous approximation (CA) to form two policies for assessment routing.
► Approximations are shown to accurately predict true policy implementation cost.
► Compared to tabu search, policy solutions are competitive in many settings.
► Policy solutions are less operationally complex than tabu solutions.