چكيده به لاتين
Health care planning, due to its sensitive and essential role in ensuring the health of society, is receiving special attention from managers and officials. On the other hand, equity in providing health services in a way that everyone can easily get the required services at any time and any place is a basic problem. In this regard, the proposed research presents a bi-objective model that follows geographical equity and equity in access. In addition to strategic decisions such as location and allocation, capacity planning has also been considered. Since the above decisions are generally taken in long-term planning, multi-period models seem to be more practical. For this purpose, the developed model is multi-period and to be more realistic, the delay time in opening a new facility, which usually occurs due to the lack of budget, has also been taken into account. Considering the time horizon in planning and the dynamic nature of the decision-making environment, the decisions that seem efficient and effective in one period may not be so in later periods, therefore, it is necessary to redesign the existing network. The concept of network redesign is incorporated into the model through policies such as opening and closing of various service providers and expanding and transferring capacity. Besides, the dynamism of the environment and the lengthening of the planning horizon intensify the effect of uncertainty on issues. To address the uncertainty, a new robust possibilistic approach based on the conditional value at risk measure (CVaR) has been developed. To solve the bi-objective model, the augmented epsilon constraint method has been used, and also to evaluate the efficiency of the model, statistical tests have been implemented based on the results of several numerical examples. The results of statistical tests indicate the significant difference between robust possibilistic II and the developed robust possibilistic approach based on CVaR. To be more precise, the mean and standard deviation of the deviations are significantly improved using the CVaR-based robust possibilistic approach compared to the robust possibilistic method (RPPII). Finally, sensitivity analysis for some parameters has been done and several managerial insights about the model have been discussed.
