چارچوبي مدل‌رانده براي مدل‌سازي و ارزيابي راهبردهاي هماهنگي براي واكنش به بحران

Coordination strategies for crisis response mainly deal with task allocation. Due to the emergency situation in emergency response environments (EREs), i.e., environments in which crises occur, and the limited time for quick and timely response, crisis managers try to use the modeling and simulation of coordination strategies before a crisis occurs to test different strategies and choose the best one. However, programming for simulating different coordination strategies is a difficult and time-consuming task for domain experts who usually do not have programming skills. To solve the problem, a framework is presented for modeling, simulating, and eva‎luating coordination strategies. The proposed framework includes three languages with tool support: (1) The CoorERE for modeling entities involved in a crisis, (2) The CoordinationML as a basic modeling language for coordination strategies, and (3) The AuctionML as an extension of CoordinationML for modeling auction-based coordination strategies which are one of the most common strategies for task allocation in EREs. Another main component of the framework is the transformation engine including model-to-code transformations for automatic code generation from designed models. The generated code can be executed on the CoorERE-designed models to generate the output model. The output model shows team formations, task allocations, resource allocations, and strategy performance. To show the applicability of the proposed framework, two EREs with different complexities and two inter and intra-organizational auction-based coordination strategies are modeled using CoorERE and AuctionML respectively. Then, the performance of the strategies is measured by the automatic generation of coordination strategies and execution of the strategies on the CoorERE models. Also, CoorERE is compared with MDD4ABMS, one of the most recent tools for modeling and simulating multi-agent systems. For this purpose, an experimental study including objective and subjective eva‎luation is conducted. Objective eva‎luation considers the required development time to fully model a case study. While the subjective eva‎luation compares two tools in terms of the extent to which they meet the requirements, and qualitative measures such as usability, productivity, and expressiveness. The eva‎luation results demonstrate that CoorERE reduces development time by 47%. Furthermore, in terms of quality criteria, CoorERE has achieved a superior score compared to MDD4ABMS.

6