| Materials distribution is an important part of manufacturing production,it plays an important role on increasing corporate profits and improving customer satisfaction by raising efficiency and reducing costs.Tools and methods of materials distribution within different enterprise are also different.The production of heavy products have complex process,sub-materials,long production time and other characteristics.In order to undertake varieties materials transport task,multi-type vehicles such as crane and forklift are often chose to distribute materials at the same time.Previously,VRP research always stick to one type of delivery vehicles,and ground transportation vehicles were more focused on,while the study of machine transportation in the air is researched less.But under the real distribution situation in the heavy product assembling workshop,crane and ground transport vehicle are different from loading ability,the target distribution and running characteristics.The systematic research on the distribution path optimization among various types of vehicles in the workshop has profound practical significance.Based on above consideration,considering the situation that diverse models of the ground delivery vehicles and crown block distribute the materials jointly in the large assembly shop at the same time,the materials vehicle routing problem is optimized to provide theoretical and practical support for the actual distribution mode in this workshop.Firstly,a large amount of analysis of the vehicle routing problem,the crane distribution research and hybrid optimization algorithm are made.The analysis shows that on the study of crane distribution problem among domestic and foreign scholars,most of the scholars who analyze the problem focus on the crane undertaken less than 30 distribution tasks,which is a contradiction status of the busy working condition.The vehicle routing optimization both in crane and ground vehicles are also studied less by predecessors.Based on that,this paper identifies the research ideas of optimization in both crane and ground transportation vehicles using hybrid optimization algorithm.Secondly,the studies on the genetic algorithm,ant colony algorithm,genetic ant colony hybrid algorithm and K-means clustering analysis are studied in detail.The analysis found that the genetic algorithm and ant colony algorithm has strong parallel and self-organization,suitable to solve complex optimization problems such as vehicle routing planling.Although each algorithm has flawed,but the advantage is complementary between genetic algorithm and ant colony algorithm,which could improve the optimization performance of the algorithm.Based on that,the basic idea of genetic ant colony hybrid algorithm adopted in the article is also analyzed in the chapter.Using the positive feedback characteristics of ant colony algorithm to optimize direction of genetic algorithm could avoid the disadvantages of genetic algorithm premature convergence and improve optimizing efficiency.On the basis of learning theory knowledge,the organizational structure,production type,plant layout,the existing distribution methods and other information of the heavy product assembling workshop C are settled.The analysis finds that problems are existing both in crane and ground vehicles materials distribution of workshop C.The dangerous distribution way,casual distribution routes with high distribution costs,low delivery efficiency and poor distribution vehicle choice are main problem of the ground vehicles.The poor delivery time of material which could not be guaranteed,the complex distribution path and the negative crane operators are the main problem of the crane distribution.Aimed at these problems,suggestions are put forward to make improvement.The plant distribution route is planned based on the improving suggestion.Based on the improving scheme,this model is established to optimize the distribution program.The objective of multiple-vehicles soft time windows vehicle routing optimization model is the minimum total cost,including fixed costs,loaded running cost,non-loaded running cost and time penalty cost respectively.Genetic algorithms and genetic ant colony hybrid algorithm are used to optimize the model.Optimization results show that genetic ant colony algorithm outperforms genetic algorithm.After optimization,the distribution cost reduces from 2502 yuan per unit to 2202 yuan per unit,which decreases by 12%.Distribution distance reduces 114m per unit,lower by 7.6%.The timeliness of delivery enhances to a large extent,premature time delivery is reduced by 32.6%,while late time delivery reduces by 81.8%.According to distribution of materials for the crane scheduling and route optimization,based on the distribution properties of transit vehicles with multi-task and the relative lack of resources in the article,K-means clustering analysis method is used to make delivery task cluster analysis according to the task needs time window.Through the analysis above,the problem is transited into six traveling salesman problem.Traveling salesman optimization model is established to solve the problem.The genetic ant colony algorithm is used to solve the traveling salesman problems.The traveling salesman problems are connected in the whole process based on each start and end time.The results show that compares with the previous case,the optimized distribution program reduces the total cost of 575 yuan per unit,saving the cost by 10.8%.11.4%delivery time delay is decreased to guarantee the delivery timeliness.The rest period of crane operator is 1.45 times than before,which could provide more adequate adjusting time.Most researches in the assembly workshop studied by scholars only focus on one or several models.This paper analyze both various types of ground transportation vehicles and crane routing optimization problems from the overall delivery perspective.Based on the multi-task characteristics of crane,the cluster analysis on the mission is made according to the delivery time window to simplify the distribution model.The conclusions has a good guiding significance on the theoretical analysis and practical application. |
PDF Full Text Request