考虑时变风险的成品油二次配送路径优化

doi:10.16085/j.issn.1000-6613.2019-1196

摘要/Abstract

摘要：

成品油属于易燃易爆炸危险品，其配送过程一旦发生事故后果严重。考虑到配送过程人群聚集程度、载重和车速的动态变化，本文建立时变风险度量模型，运用BP神经网络预测车速，预测了城市高速路段和主干路路段的车速。在此基础上，构建了同时考虑配送成本与时变风险的成品油二次配送双目标路径优化模型，并提出改进的多目标列队竞争算法进行求解。本文还采用改进的多目标列队竞争算法求解文献报道的实例，求得非支配解完全支配文献报道中的非支配解，验证了算法的有效性。某武汉石化企业实例研究表明，预测的车速具有较高的准确度，所提出的方法能同步考虑配送过程的时变风险和配送成本，求得的Pareto最优配送路径集可以为企业决策者提供多样化的选择方案，具有较好的应用价值。

关键词: 时变风险, 成品油配送, 预测, 优化, 算法

Abstract:

Refined oil belongs to inflammable and explosive dangerous products. Once an accident occurs in the distribution process, the consequences are serious. Considering the dynamic changes of population aggregation, load and speed in distribution process, a time-dependent risk measurement model was established. The BP neural network was used to predict vehicle speed, and the speed of expressway and main road in Wuhan was predicted. On this basis, a bi-objective route optimization model of refined oil secondary distribution route considering both distribution cost and time-dependent risk was proposed, and an improved multi-objective line-up competition algorithm was proposed to solve the problem. The improved multi-objective line-up competition algorithm was used to solve the examples reported in the literature, and the obtained non-dominated solutions completely dominated the non-dominated solutions in the literature reports, which verified the effectiveness of the algorithm. A case study of Wuhan petrochemical enterprise showed that the predicted speed had a high accuracy, the proposed method could simultaneously consider the time-dependent risks and distribution costs in the distribution process, and the obtained Pareto optimal distribution routing set could provide diversified options for enterprise decision-makers, and had good application value.

Key words: time-dependent risk, distribution of refined oil, prediction, optimization, algorithm

中图分类号:

TQ015.9

韦金银,郭琪,史彬,鄢烈祥. 考虑时变风险的成品油二次配送路径优化[J]. 化工进展, 2020, 39(4): 1597-1604.

Jinyin WEI,Qi GUO,Bin SHI,Liexiang YAN. Route optimizing on secondary distribution of refined oil considering time-dependent risk[J]. Chemical Industry and Engineering Progress, 2020, 39(4): 1597-1604.

图/表 12

参考文献 19

1	杨路燕.基于风险分析的成品油二次配送路径优化研究[D]. 大连: 大连海事大学, 2014.YANG L Y. Route optimization for the sencondary distribution of oil product on the basis of risk analysis[D]. Dalian: Dalian Maritime University, 2014.
2	吕圣杰.我国成品油二次配送路径优化分析[J]. 中国储运, 2016(2): 118-120.
	LÜ S J. Analysis on optimization of secondary distribution route of refined oil in China[J]. China Storage & Transport, 2016(2):118-120.
3	CORNILLIER F, BOCTOR F, RENAUD J. Heuristics for the multi-depot petrol station replenishment problem with time windows[J]. European Journal of Operational Research, 2012, 220(2): 361-369.
4	AVELLA P, BOCCIA M, SFORZA A. Solving a fuel delivery problem by heuristic and exact approaches[J]. European Journal of Operational Research, 2004, 152(1): 170-179.
5	仝自强,李鹏翔.考虑实时路况和车辆周转率的成品油配送路径优化研究[J]. 工业工程与管理, 2019, 24(2): 109-115.
	TONG Z Q, LI P X. Vehicle routing problem of refined oil distribution considering real-time traffic condition and vehicle turnover rate[J]. Industrial Engineering and Management, 2019, 24(2): 109-115.
6	孙丽君, 石海洋, 胡祥培. 考虑司机工作量均衡的成品油配送优化[J]. 系统工程理论与实践, 2018, 38(3): 677-686.
	SUN L J, SHI H Y, HU X P. An optimization method of product oil distribution considering drivers workload balance[J]. Systems Engineering: Theory & Practice, 2018, 38(3):677-686.
7	张文博, 苏秦, 程光路. 基于动态需求的带时间窗的车辆路径问题[J]. 工业工程与管理, 2016, 21(6): 68-74.
	ZHANG W B, SU Q, CHENG G L. Vehicle routing problem with time windows based on dynamic demands[J]. Industrial Engineering and Management, 2016, 21(6):68-74.
8	葛显龙,张慧.城市实时交通路网车辆路径优化问题研究[J]. 工业工程与管理, 2018, 23(3): 140-149, 156.
	GE X L, ZHANG H. Study on the optimization of vehicle routing problem in urban real time traffic network[J]. Industrial Engineering and Management, 23(3): 140-149, 156.
9	王旭坪, 韩进军, 董杰. 考虑脆弱点的成品油二次配送风险层级优化[J]. 管理学报, 2017, 14(5): 751-758.
	WANG X P, HAN J J, DONG J. Risk layer-optimization for the secondary distribution of refined oil considering vulnerable centers[J]. Chinese Journal of Management, 2017, 14(5): 751-758.
10	袁文燕, 徐腾飞, 杨丰梅. 基于新风险度量方式的危险品车辆路径双目标优化模型[J]. 数学的实践与认识, 2016, 46(14): 275-284.
	YUAN W Y, XU T F, YANG F M. Bi-objective decision model for vehicle routing of hazardous material based on new risk measure[J]. Mathematics in Practice and Theory, 2016, 46(14): 275-284.
11	王录仓. 基于百度热力图的武汉市主城区城市人群聚集时空特征[J]. 西部人居环境学刊, 2018, 33(2): 52-56.
	WANG L C. Spatial-temporal characteristics of urban population aggregation based on Baidu heat map in central areas of Wuhan city[J]. Journal of Human Settlements in West China, 2018, 33(2): 52-56.
12	SHI B, YAN L X, Wu W. Rule-based scheduling of single-stage multiproduct batch plants with parallel units[J]. Industrial & Engineering Chemistry Research, 2012, 51(25): 8535-8549.
13	SHI B, QIAN X R, SUN S S, et al.Rule-based scheduling of multi-stage multi-product batch plants with parallel units[J]. Chinese Journal of Chemical Engineering,2017, 25(8): 1022-1036.
14	何晓旭, 钱欣瑞, 鄢烈祥, 等. 热集成变压精馏分离乙二胺-水共沸体系的模拟优化[J]. 化工进展, 2018, 37(6): 2426-2431.
	HE X X, QIAN X R, YAN L X, et al. Simulation and optimization for azeotrope of ethylenediamine-water based on heat-integrated pressure-swing distillation[J]. Chemical Industry and Engineering Progress, 2018, 37(6): 2426-2431.
15	刘海燕, 潘欣, 史彬, 等. 基于LCA与NSGA-II的混合算法解多目标优化问题[J]. 计算机与应用化学, 2014, 31(12): 1461-1466.
	LIU H Y, PAN X, SHI B, et al. A hybrid algorithm based on LCA and NSGA-II solving multi-objective optimization problem[J]. Computers and Applied Chemistry, 2014, 31(12): 1461-1466.
16	王佩, 史彬, 鄢烈祥. 炼油厂原油选购多目标优化[J]. 大连理工大学学报, 2019, 59(3): 251-256.
	WANG P, SHI B, YAN L X. Multi-objective optimization of crude oil purchase in refinery[J]. Journal of Dalian University of Technology, 2019, 59(3): 251-256.
17	姚晓晖, 王山, 汪彤. 危险化学品公路运输事故原因及发生频率研究[J]. 安全, 2005, 26(5): 37-38.
	YAO X H, WANG S, WANG T. Study on causes and frequency of road transportation accidents of dangerous chemicals[J]. Safety & Security, 2005, 26(5): 37-38.
18	黄冀卓, 王湛. 基于遗传算法的抗震钢框架多目标优化设计[J]. 力学学报, 2007, 39(3): 389-397.
	HUANG J Z, WANG Z. Multi-objective optimization design of aseismic steel frames using genetic algorithm[J]. Chinese Journal of Theoretical and Applied Mechanics, 2007, 39(3): 389-397.
19	许欢. 多目标进化算法在物流配送车辆路径问题中的应用研究[D]. 广州: 广东工业大学, 2013.XU H. Research of multi-object evolutionary algorithm applications on the vehicle routing problem in logistics[D]. Guangzhou: Guangdong University of Technology, 2013.

序号	经度	纬度	需求量/t	最早接受服务时间/min	最晚接受服务时间/min	服务时间/min
0	114.468527	30.658865	0	0	660	0
1	114.194116	30.600468	2	620	660	30
2	114.340954	30.660551	3	180	220	30
3	114.321492	30.560212	3.5	90	130	30
4	114.223965	30.522685	4	440	480	30
5	114.292797	30.501221	1.5	290	330	30
6	114.371355	30.466157	2.5	210	250	30
7	114.444598	30.506969	3	510	550	30
8	114.353382	30.599453	4.5	380	420	30
9	114.613348	30.493312	1	270	310	30
10	114.611495	30.618218	2.5	110	150	30
11	114.601149	30.649721	2.5	590	630	30
12	114.227782	30.516472	3	570	610	30
13	114.421778	30.635173	1.5	410	440	30
14	114.209086	30.578514	2	40	80	30
15	114.210273	30.466125	3.5	300	340	30
16	114.528845	30.550224	1.5	120	160	30

序号	经度	纬度	需求量/t	最早接受服务时间/min	最晚接受服务时间/min	服务时间/min
0	114.468527	30.658865	0	0	660	0
1	114.194116	30.600468	2	620	660	30
2	114.340954	30.660551	3	180	220	30
3	114.321492	30.560212	3.5	90	130	30
4	114.223965	30.522685	4	440	480	30
5	114.292797	30.501221	1.5	290	330	30
6	114.371355	30.466157	2.5	210	250	30
7	114.444598	30.506969	3	510	550	30
8	114.353382	30.599453	4.5	380	420	30
9	114.613348	30.493312	1	270	310	30
10	114.611495	30.618218	2.5	110	150	30
11	114.601149	30.649721	2.5	590	630	30
12	114.227782	30.516472	3	570	610	30
13	114.421778	30.635173	1.5	410	440	30
14	114.209086	30.578514	2	40	80	30
15	114.210273	30.466125	3.5	300	340	30
16	114.528845	30.550224	1.5	120	160	30

时间段	快速路/km?h^-1	主干路/km?h^-1
7:30—9:00	40.5478	36.8785
9:00—10:30	56.5088	40.4119
10:30—12:30	65.2925	43.3636
12:30—14:30	66.4418	44.0517
14:30—17:00	64.5524	40.2875
17:00—19:00	50.3678	32.1878
19:00—20:00	67.2248	43.3285

时间段	快速路/km?h^-1	主干路/km?h^-1
7:30—9:00	40.5478	36.8785
9:00—10:30	56.5088	40.4119
10:30—12:30	65.2925	43.3636
12:30—14:30	66.4418	44.0517
14:30—17:00	64.5524	40.2875
17:00—19:00	50.3678	32.1878
19:00—20:00	67.2248	43.3285

路径编号	经过加油站点顺序
1	13 - 14 - 6
2	2 - 5 - 15
3	7 - 11
4	8 - 3
5	16 - 9 - 10 - 12
6	4 - 1