In this paper, we consider the process of collection of deteriorating goods from intermediate storage locations and their delivery to a processing plant of a storage facility within a given period. During this process, the collection vehicles make multiple trips. Since the quality of products decays in time, it is necessary to find the optimal scheduling of collection vehicles and determine the optimal sequence of visiting the storage locations, as well as the moments of visits, so that the total quality of collected goods is maximized. This optimization problem is defined as a mixed-integer nonlinear program with continuous-time, nonlinear objective function and quadratic and linear constraints, which simultaneously maximizes the quality of collected goods and minimizes the fleet size of collection vehicles. To solve the problem, we propose a simulated annealing (SA) based heuristic approach, while to cope with uncertainty and provide realistic data for rescheduling and improving the deterministic scheduling plan, we develop a simulation model of a real collection process. The proposed optimization and simulation approach, which is based on successive repeated runs of the SA algorithm and the simulation model, allows the improvement of the collection process by rescheduling the vehicles. At the same time, it allows us to consider the problem as a dynamic and stochastic vehicle scheduling problem. The expected effects and applicability of the proposed approach are tested on numerical examples.

在本文中，我们考虑了在一定时期内从中间存储地点收集恶化的货物并将其运送到存储设施的加工厂的过程。在此过程中，收集车进行多次行程。由于产品的质量会随着时间的推移而下降，因此有必要找到收集车辆的最佳计划，并确定访问存储地点的最佳顺序以及访问的时间，以使所收集货物的总质量最大化。该优化问题定义为具有连续时间，非线性目标函数以及二次和线性约束的混合整数非线性程序，该程序同时使所收集货物的质量最大化，并使所收集车辆的车队规模最小。为了解决这个问题 我们提出了一种基于模拟退火（SA）的启发式方法，同时为了应对不确定性并为重新计划和改进确定性计划计划提供现实的数据，我们开发了一个实际收集过程的模拟模型。所提出的优化和仿真方法基于SA算法和仿真模型的连续重复运行，可以通过重新安排车辆的时间来改进收集过程。同时，它使我们可以将问题视为动态且随机的车辆调度问题。在数值实例上测试了该方法的预期效果和适用性。我们开发了一个真实收集过程的仿真模型。所提出的优化和仿真方法基于SA算法和仿真模型的连续重复运行，可以通过重新安排车辆的时间来改进收集过程。同时，它使我们可以将问题视为动态且随机的车辆调度问题。在数值实例上测试了该方法的预期效果和适用性。我们开发了一个真实收集过程的仿真模型。所提出的优化和仿真方法基于SA算法和仿真模型的连续重复运行，可以通过重新安排车辆的时间来改进收集过程。同时，它使我们可以将问题视为动态且随机的车辆调度问题。在数值实例上测试了该方法的预期效果和适用性。它使我们可以将问题视为动态的随机车辆调度问题。在数值实例上测试了该方法的预期效果和适用性。它使我们可以将问题视为动态的随机车辆调度问题。在数值实例上测试了该方法的预期效果和适用性。