School of Innovation and Entrepreneurship, Shandong Women's University, Jinan, Shandong 250300, China.
Comput Intell Neurosci. 2022 Sep 12;2022:5242251. doi: 10.1155/2022/5242251. eCollection 2022.
Laboratory equipment planning is a very important task in modern enterprise management. Laboratory equipment planning by computer algorithm is a very complex NP-hard combinatorial optimization problem, so it is impossible to find an accurate algorithm in polynomial time. In this study, an improved genetic algorithm is used to solve and analyze the comprehensive planning of laboratory equipment. After analyzing the traditional heuristic algorithm and genetic algorithm to solve the simple laboratory equipment planning problem, the simple laboratory equipment planning is designed and implemented according to the principle of the heuristic algorithm. Finally, the algorithm is implemented in Python. A general equipment planning based on genetic algorithm with two selection operators is realized. Two constraints of test start and completion time are given. In the scenario of using multiple test equipment for a test project, the possible solutions of laboratory equipment planning under given constraints are analyzed. The efficiency coefficient is not necessarily a constant, it is related to the output characteristics of energy equipment. Three independent planning algorithms are used to solve the actual test requirements. One is the planning method based on manual test scheduling in the test cycle of experimental instruments, the other is the genetic algorithm for gene location crossover operator, and the third is the genetic algorithm for experimental part crossover operator. The planning solutions obtained by the three algorithms are compared from three aspects: the shortest time to complete the test, the calculation time of the algorithm, and the utilization of the test equipment.
实验室设备规划是现代企业管理中非常重要的任务。通过计算机算法进行实验室设备规划是一个非常复杂的 NP 难组合优化问题，因此不可能在多项式时间内找到准确的算法。在本研究中，使用改进的遗传算法来解决和分析实验室设备的综合规划。在分析了传统启发式算法和遗传算法来解决简单的实验室设备规划问题之后，根据启发式算法的原理设计并实现了简单的实验室设备规划。最后，在 Python 中实现了算法。实现了一个具有两个选择算子的基于遗传算法的通用设备规划。给出了两个测试开始和完成时间的约束。在使用多个测试设备进行测试项目的情况下，分析了给定约束下实验室设备规划的可能解决方案。效率系数不一定是常数，它与能源设备的输出特性有关。使用三种独立的规划算法来解决实际测试需求。一种是基于实验仪器测试周期中手动测试调度的规划方法，另一种是基于基因定位交叉算子的遗传算法，第三种是基于实验部分交叉算子的遗传算法。从完成测试的最短时间、算法的计算时间和测试设备的利用率三个方面对三种算法得到的规划解决方案进行了比较。
