仓配装一体化产品是企业为客户提供一站式仓储、配送和安装服务的一种新型产品，并逐渐成为新的消费趋势。然而，随着中国经济增长的放缓，产业经过快速扩张后正处于调整期，市场已从供少于求转变为供过于求，企业在供给侧改革中也面临巨大的生存压力。另一方面，在成本上涨、利润下降的商业环境下，企业之间的竞争也愈发激烈。因此，如何充分利用企业现有的资源，降低企业的运营成本也成为行业内每个企业需要思考的问题。在此背景下，众多企业从传统的商业模式向智能制造转型，在生产、仓储和配送等环节发生深刻变革，并开始为客户提供一站式仓储、配送和安装服务。其中，家居产品是仓配装一体化产品的典型代表。通过对家居行业龙头企业A公司的实地调研，我们以家居产品为研究对象，深入分析如何优化和改善企业现有的生产计划、仓库调度策略和配送服务，以降低A公司的运营成本。本文的主要内容概括如下： (1)针对生产环节，研究了如何设计合理的产品生产与产品交付计划，包括(i)确定产品的加工顺序，(ii)确定产品交付到顾客的时间和方式。该问题的目标是最小化产品延误成本、库存成本和交付成本的加权之和。我们主要考虑了顾客数量任意和固定情况下对应问题的计算复杂度，证明了当顾客数量任意时，此研究问题是强NP-hard的，而当顾客数量固定时，此问题是一般NP-hard的。同时，针对顾客数量固定，我们在产品的规格大小相同、产品的加工时间相同和交付选用的交通工具运输能力无限制这三类特殊的情形下提出了伪多项式或多项式时间精确算法。同时，我们运用了企业实际数据，将算法与企业实际应用的“先制定生产计划再制定交付计划”的经验式决策方法作对比，数值结果显示算法可有效地降低企业约11％左右的成本。 (2)针对仓储环节，研究了在高效率的自动化仓库中产品出入库的调度问题。在生产环节中，半成品或者成品需要暂时存放在仓库中。针对一种在货架的底层有多个出口位置以供拣选人员做拣选任务新型自动化仓储系统，管理人员可以借助此仓储系统将存储和分拣任务统一于一处，既提高了空间利用率，又降低了能源消耗。我们聚焦于出入库任务调度和出口位置分配问题，需要同时决策出入库任务的顺序以及出库任务与出口位置的匹配，目标是最小化堆垛机完成所有任务的总移动距离。我们建立了求解该问题的混合整数规划模型，并将问题分解成两个子问题：(i)确定出入库任务的顺序，(ii)在给定出入库顺序的情况下决定出库货物释放到哪一个出库位置。其中，后一个子问题可以在多项式时间内求出最优解。根据此性质我们设计了基于遗传算法和指派问题的两阶段启发式算法。我们先通过产生随机的数据集来分析算法的性能，最后利用企业实际数据，与企业实际应用的“先到先服务”的经验式决策方法作对比，数值实验的结果表明了该模型和算法的有效性，降低了超过20％的运作时间。 (3)针对配送环节，以A公司的配送和维护服务为背景，研究了产品配送服务中技术人员路线选择和调度的问题，解决企业在将产品交付到不同地理位置分销商或顾客时所面临的问题。我们从三个方面确定最优调度计划：(i)技术人员与团队的分配方案，(ii)在满足任务请求和团队资质匹配的情况下团队与任务的分配方案，(iii)团队执行各自任务的路线。通过此计划要实现的目标是最小化运营成本，包括旅行成本，违反时间窗口的惩罚成本和外包成本。我们结合自适应大邻域搜索算法和禁忌搜索算法来寻找初始可行解，并且提出了一种基于拉格朗日松弛的启发式算法。我们先通过随机产生数据集来分析算法的性能，然后利用企业实际数据，与企业实际应用的“对任务聚类，用最近邻算法确定路线”的经验式决策方法作对比，数值实验的结果表明了此算法能在合理的计算时间内找到高质量的解，且有效地降低约20％的总行驶距离。 本研究融合整数规划理论、调度方法以及组合优化理论，对家居企业在生产-仓储-配送三方面的运作问题进行了探索，提供了新的理论方法，研究成果有利于丰富生产、仓储和配送方面的理论与方法。从应用价值上，本研究成果帮助A公司解决产品在生产、仓储和配送等环节中遇到的难题，在不影响顾客服务质量的同时有效地降低企业成本并创造利润空间。从长远角度看，本研究成果既能有助于企业信誉的提升，也能为企业运营能力状况提供定量评估，同时也为其他仓配装一体化产品企业提供借鉴和参考，保障和促进我国相关产业的发展。 关键词：生产计划，出入库调度，配送问题，启发式算法

 Products whose purchase combines warehousing and delivery are a new offering of one-stop warehousing, distribution, and installation service companies, and such products have gradually become a new commerce trend. However, as China’s economic growth slows down, the industry is undergoing an adjustment period after rapid expansion. The market has changed from a supply-demand to an oversupply market, and enterprises are facing tremendous pressure to survive this supply-side reform. Also, in the business environment of rising costs and falling profits, competition among enterprises has become fiercer. How to make full and reasonable use of the company's existing resources and reduce the operational costs of enterprises has become a problem that every enterprise in the industry needs to solve. In this context, many companies have transformed from traditional business models to smart manufacturing, and have undergone profound changes in production, warehousing, and distribution, providing customers with one-stop warehousing, distribution, and installation services. Household products are typical representatives of such warehouse integrated products. Through field research on the leading company A in the home furnishing industry, we use household products as research objects to study how to optimize and improve the company’s existing production plans, warehouse scheduling strategies, and distribution services to reduce the operational costs for company A. The contents of this thesis are summarized as follows: This thesis first considers the production of household products, and studies how to design a reasonable production and delivery plan, including (1) determining the processing order of household products, and (2) determining the time and manner of delivery of household products to customers. The goal of this solution is to minimize the weighted sum of product delays, inventory costs, and delivery costs. We prove that when the number of customers is arbitrary, the question is NP-hard in the strong sense, and when the number of customers is fixed, the problem is NP-hard in the ordinary sense. Meanwhile, for a fixed number of customers, we propose a pseudo-polynomial or polynomial time algorithm for the three special cases of the same product size, the same processing time, and the unlimited transport capacity of the selected vehicles. At the same time, we also use the actual data of the enterprise to compare the performance of the proposed algorithm with the experiencial decision-making method of “first making the production plan and then making the delivery plan”. The numerical results show that the algorithm can effectively reduce the total cost by 11 %. The study then considers the storage and retrieval problem for household products. After the production process, semi-finished products need to be temporarily stored in a warehouse. The automated storage and retrieval system with multiple in-the-aisle pick positions is a new warehousing technology which combines the storage and order picking processes, which will be adopted by company A. The typical feature of this system is that there exist multiple picking locations at the bottom of the rack for workers. This thesis focuses on the order sequencing problem, which will be jointly optimized by choosing a picking position for each retrieval command. A mixed integer programming model with the objective of minimizing the total travel distance of the crane to complete all commands is built, and a two-stage heuristic algorithm is designed to solve this problem. We first analyze the performance of the algorithm by randomly generating the dataset, and finally use the actual data of the enterprise to compare the proposed algorithm with the popular “first-come, first-serve” policy implemented by the enterprise. Our numerical experiments show the effectiveness of the new model and algorithm, which offers a cost reduction of about 20%. Finally, we consider the distribution problem of household products, studying the routing and scheduling problems of multi-warehouse technicians under constraints of soft time windows and lunch break demands. The optimal scheduling plan is determined from three aspects: (1) the allocation of the technical staff team; (2) the arrangement of the team in the case of satisfying the customer request and the team qualification; and (3) the route for the team to perform their tasks. The objective is to minimize the total operational costs, including transfer costs, soft time window violations, and outsourced service costs. We combine an adaptive large neighborhood search algorithm and a tabu search algorithm to find the initial feasible solution and then propose a Lagrangian-based heuristic algorithm. We first analyze the performance of the algorithm by randomly generating datasets, and then use the actual data of the enterprise to compare the new algorithm with the artificial decision-making method of “clustering the task and determining the route with the shortest neighbor algorithm”. The results of numerical experiments show that the algorithm can find high-quality solutions within a reasonable calculation time and effectively reduce the total operational costs by more than 20%. This study combines integer programming theory, scheduling methodology, and combination optimization theory to explore the three aspects of household product production-warehousing-distribution, providing new theoretical methods. The research results are conducive to enriching the theory of production, storage, and distribution methods. For practice, the research results help company A solve the problems encountered in its production, warehousing, and distribution of household products, effectively reducing the cost of the enterprise and creating profit margins without affecting the quality of customer service. In the long run, such practices can not only improve the credibility of the enterprise but also provide a quantitative assessment of the operational capabilities of the enterprise. At the same time, it also provides a reference for other household products companies to guarantee and promote the development of related industries in China. Key Words: Production plan; Outbound and inbound scheduling; Delivery problem; Heuristic