Optimization Methods For Order Picking Under Vegetables B2C Online Direct Selling Mode

China has been the largest e-commerce country with over 10 trillion transactions since 2013.People live on food and three meals a day are the guarantee of human survival.The daily purchase of fresh food has turn to the Internet,and fresh e-commerce is praised as "the last blue sea of retail e-commerce " because of the huge sales scale and frequent purchase.At present,there are many fresh e-commerce market modes and superiority of fresh B2C online direct selling e-commerce mode is obvious among many modes.Fresh vegetables are picked directly from the production base,through cleaning,precooling,minimum sales unit packaging,picking and packing in the the distribution,the standard fresh products are distributed directly to the customers.This mode directly connects farmers and customers and reduces many intermediate links and costs of agricultural supply chain.It's a new way of distributing fresh food.The picking operation of fresh B2C direct selling mode is the most core operation in the distribution center and is the bottleneck of the whole operation of distribution center no matter from the perspective of cost and efficiency.The current picking operation in distribution center is disordered and lack of scientific guidance.Its resource utilization is very low.The low efficiency and high cost of order picking operation in distribution center lead to the disconnection of upstream and downstream operations,a large number of fresh rotten and higher warehouse buffers.Orders can not be distributed on time which finally directly affects customers' three meals a day which is hard to retain customers.With the increasing sales volume,greater challenges are faced by B2C online direct selling mode.Many papers on picking operations optimization have been published in the international mainstream frontier journals during nearly four years.Therefore,the research on the picking operation optimization has important academic and practical significance.Based on the national project support and the actual survey of enterprises,this paper proposed the JIT-based order picking operation mode.The order picking operation of fresh e-commerce has the following characteristics.Firstly,the fresh distribution requirements of "half day delivery" and "next day delivery" make the very high timeliness of order picking.Secondly,fresh food is seasonal and perishable and cannot be stored in advance,so that the supply of picking goods is uncertain.Fluctuated price,holidays and other factors make the demand for picking goods uncertain.The picking system in this paper is the human based picking mode and the picking operation has the flexible cycle time.The above characteristics of picking operations lead to a high uncertainty in picking resource allocation and batch picking scheduling.Meanwhile,the optimization of resource allocation at the strategic level and the order picking scheduling in the operation level interact with each other.Resource optimization configuration is the core element that affects order picking scheduling.Therefore,how to study the picking operation optimization under uncertainty which different decision levels interact each other is the core problem in this paper.This problem is NP hard problem,with high uncertainty,cross-level,multi-stage,multi-periods and unstructured feature.These difficulties make the traditional research results unable to solve the order picking operation optimization problem for fresh e-commerce.Considering exist research lacking vegetable picking mode design and the study of the picking resource and scheduling on the background of direct-selling e-commerce mode,based on sequential decision-making idea and stochastic programming theory,this paper depicts the uncertain factors in picking operation optimization and puts forword two-stage decomposed decision which includes different decision levels including resource optimization allocation and daily order picking scheduling.These two subproblems,resource optimization allocation and order picking scheduling are relatively independent and influence each other.At the same time,based on exact algorithm,intelligent algorithm as well as batch scheduling,a two-tage decomposition method is proposed for resource optimization problem which combines the virtues of exact algorithm and intelligent algorithm and two-stage solving scheme is presented for batch picking scheduling problem which combines qualitative and quantitative decision rules.The influence of picking scheduling on the decision is considered in the process of resource optimization configuration,and the batch picking scheduling is solved on the given picking resource allocation to realize the overall decision.The methods in this paper,on one hand,can solve the complex multi-stage decision-making problem with multi-level decision-making interaction;on the other hand,it can provide decision-making support for the resource allocation of order picking and daily order batching scheduling of e-commerce enterprises.The main research work of this paper is as follows.(1)Overview of vegetables B2C online direct selling mode,definition of research problems in this paper and decision process analysis of the picking operations optimization.The strategies and methods of vegetable picking operation are summarized,and the characteristics and complexity of vegetable picking operation optimization are analyzed.To define the research scope and decision-making object of picking operation optimization,this paper analyzes the uncertain factors in the optimization process of picking operations and analyzes the principle of decision-making process.Lastly,this paper proposes a two-stage decision-making method for the picking operations optimization,and then discusses the characteristics of the two sub-problems involved.(2)Two-stage stochasitc programming method for resource optimization on vegetables B2C online direct selling mode.A two-stage stochastic programming model is established to solve the resources allocation problem considering order picking operation scheduling which has uncertainty of the batch quantity and batch processing time.Aiming at the characteristics of NP hard,slow solving speed and non-relatively complete recource of model,a decomposition method is presented which combines the virtue of precise algorithm with the intelligent algorithm.(3)Batch picking operation scheduling method for vegetables B2C online direct selling mode.When the resource allocation is given,considering the influence of resource fatigue effect on the operation efficiency,a batch order scheduling model with decreasing nonlinearly over time operation efficiency is established,and the operation efficiency is characterized on the basis of logical equation.According to the special characteristics of the vegetable orders,the batching criterion and batching method based on the similarity rule is put forward.Finally,based on the sequential decision theory,a two-stage solution is proposed to solve the batch order scheduling problem and corresponding heuristic algorithms are developed for the order batching and batch order scheduling.(4)Instance verification.Combined with the actual situation of the cooperative enterprise,this paper simulates the actual operational data.For picking resource optimization method and batching picking scheduling method proposed in this paper,numerical experiment analysis is conducted.The results show that the effectiveness of the proposed method and algorithm,and corresponding management highlights are put forward.Focused on the complex,multi-stage,unstructured and uncertain picking operation optimization for vegetables and fruits B2C e-commerce mode which has different interactional levels of decision problems,based on the theory of operational research,stochastic programming,production scheduling,group theory and sequential decision-making thought,order picking optimization problem is decomposed into two interacted and relatively independent subproblems which not only provides a solution scheme for multistage problem having tactics level and operation level decisions,but also expand the batch job scheduling and resource allocation optimization research theory.In the practical aspact,the conclusion of this paper provides the support method for picking operating decision of fresh e-commerce distribution center and theory basis for solving the current chaotic,unscientific order picking operation.Our methods could reduce the resource cost and improve the picking efficiency in the distribution center which has great realistic significance for the vigorous development of fresh B2C direct sales mode.