No abstract

Purpose Previously use of drones as a relief distribution vehicle was studied in several studies where required number of drones and the best locations for the relief centers were investigated. The maximum travel distance of drones without a need to recharge is limited by their endurance. Recharge stations can be used to extend the coverage area of the drones. The purpose of this paper is to find the best topology for both relief centers and recharge stations to cover a large-scale area with minimum and feasible incurred costs and waiting times. Design/methodology/approach A multi-level facility location problem (FLP) is utilized to find the optimum number of relief centers and refuel stations and their locations. It is supposed that the demand occurs according to Poisson distribution. The allocation of the demand is based on nearest neighborhood method. A hybrid genetic algorithm is proposed to solve the model. The performance of the algorithm is examined through a case study. Findings The proposed method delivers increased efficiency and responsiveness of the humanitarian relief system. The coverage area of the drones is extended by refuel stations, total costs of the system are reduced and the time to respond an emergency, which is an important factor in survival rate, is significantly decreased. Originality/value This study proposes a multi-level FLP to simultaneously account for recharge stations, relief centers and the number of required drones to cover all the demand for relief in a post-disaster period.

No abstract

Online retailers invest an enormous amount of funds in delivering products to customers. In recent years, these delivery costs have increased as a result of changes in fuel costs, which has brought new challenges to retailers in terms of offering competitive prices. Many retailers have begun to utilize a drone-based aerial delivery system as an alternative solution to overcome the problems related to the high transportation costs and traffic jams in large cities. This study provides a mathematical model for minimizing the total costs of the aerial delivery system concerned with refuel stations, warehouses, drone procurement, and transportation. The waiting time of the customers is restricted based on the M/G/K queueing system. The fuel stations and warehouses are the main components of the network. The demand (occurring at the lowest level) is ultimately satisfied via launch stations (the network's highest level). Refuel stations support drones along their long routes between the launch stations and demand points. To account for the different levels of the facilities, a multi-level facility location approach is utilized. Moreover, the nondeterministic nature of the problem is tackled using fuzzy variables. The ultimate mathematical model is a congested fuzzy capacitated multi-level facility location problem that is solved by the possibilistic approach.

Purpose The purpose of this paper is to deal with the transportation of a high number of injured people after a disaster in a highly populated large area. Each patient should be delivered to the hospital before the specific deadline to survive. The objective of the study is to maximize the survival rate of patients by proper assignment of existing emergency vehicles to hospitals and efficient generation of vehicle routes. Design/methodology/approach The concepts of non-fixed multiple depot pickup and delivery vehicle routing problem (MDPDVRP) is utilized to capture an image of the problem encountered in real life. Due to NP-hardness of the problem, a hybrid genetic algorithm (GA) is proposed as the solution method. The performance of the developed algorithm is investigated through a case study. Findings The proposed hybrid model outperforms the traditional GA and also is significantly superior compared to the nearest neighbor assignment. The required time for running the algorithm on a large-scale problem fits well into emergency distribution and the promptness required for humanitarian relief systems. Originality/value This paper investigates the efficient assignment of emergency vehicles to patients and their routing in a way that is most appropriate for the problem at hand.

In recent years, small unmanned aerial vehicles have been used to deliver medicine and goods as a solution to severe traffic jams and to serve the purpose of fast and effective delivery, especially for medical and emergency applications where time is vital. On the other hand, in the competitive market of today, retailers are considering the use of drones to minimize the customers' waiting times and as a way to lower their transportation costs. This study aims to develop a biobjective mathematical model to account for the optimum number and spatial location of facilities among a set of candidate locations such that the total travel distance, costs, and lost demand are minimized simultaneously. It is assumed that the demand occurrence follows a Poisson distribution and is uniformly distributed along the network edges. The proposed biobjective capacitated facility location model is NP-hard, thus nondominated sorting genetic algorithm II and reference-point based nondominated sorting genetic algorithm are applied to solve the problem. The performance of the algorithms, quality of solutions, and the results are investigated and discussed.

Following a large-scale disaster in a highly populated city, one can expect a large number of injured people to need urgent operation. If it can be assumed that the city in question is likely to experience a disaster (although the time of its occurrence is unknown), pre-disaster measures should be taken to mitigate the consequences of the disaster. This study aims to minimize the fatality rate through the proper assignment of operating room personnel to hospitals, such that both the expected value of the functioning operating rooms and the expected value of the service level are maximized. Service level is defined as a variable that has a negative relationship to the distance a patient is expected to travel in order to receive the required medical services. The probability of survival, for both personnel and operating rooms, depends on several stochastic factors. A bi-objective mixed integer nonlinear model is proposed for the problem. A simulated annealing algorithm, particle swarm and a genetic algorithm are developed to solve the model using a weighted metric method. The model is applied to a possible worst-case earthquake scenario in Tehran. The results show that the proposed approach significantly improves the performance of post-disaster emergency relief.

No abstract