【作者】 李晓慧；

【导师】 程鸿群；

【作者基本信息】 武汉大学 ， 管理科学与工程， 2023， 硕士

【摘要】 随着劳动力成本的上升以及日益突出的资源、环境问题,建筑业亟待向可持续发展和工业化建筑转变。装配式建筑可以缩短施工时间,减少能耗,提高施工质量,减少污染和劳动力。尽管装配式建筑有很多优点,但装配式建筑的发展未达预期,主要是因为装配式建筑建造过程中存在调度复杂、一体化管理水平不足、不确定事件频发等问题。因此,本文从装配式建筑构件供应商的角度出发,建立确定条件和不确定条件下基于安装时间窗的装配式建筑生产-运输集成调度模型,该研究旨在提升装配式建筑资源配置效率,并提高装配式建筑建造过程中应对不确定事件的能力。本文研究了确定条件下和不确定条件下装配式建筑生产-运输集成调度优化问题。在确定条件下,建立了总成本最小化目标函数。在生产阶段中,运用传统置换流水车间调度模型,并考虑工序中断、并行、非工作时间等因素。在运输阶段中,考虑到构件体积大,安装空间有限,运输车辆载重远小于订单量等因素,将运输过程建模为JIT准时交付和分批运输交付。设计了基于构件生产顺序、运输次序的分段染色体,通过自适应贪婪禁忌遗传算法(Adaptive Greedy Tabu Genetic Algorithm,简称AGTGA)中的自适应交叉和变异、破坏、禁忌、重建过程来避免陷入遗传算法局部最优解,较好地平衡了局部和全局搜索,降低无效搜索次数。并通过田口实验确定了AGTGA的最优参数值组合,利用27个算例检验了AGTGA的有效性和优越性。为了避免单个流水线生产力不足,进一步丰富算法的应用场景,研究了在不确定条件下单供应商多流水线的生产模式。运用三角模糊数表示生产加工时间和运输时间的不确定性,运用梯形模糊数表示安装时间窗的不确定性。并基于多条流水线决策之间的层次关系,建立订单流水线分配和单流水线生产-运输集成调度的双层优化模型,采用基于自适应贪婪禁忌遗传算法的双层优化求解。在上层优化中,为构件订单分配流水线,计算多流水线总成本最低的订单流水线分配方案。在下层优化中,引入考虑决策者风险偏好以及抗风险能力的模糊悲观满意度,计算单流水线下模糊悲观满意度最大的构件生产和运输顺序方案。运用案例检验了基于AGTGA的双层优化求解不确定条件下装配式建筑生产-运输集成调度的有效性。并进一步分析了确定条件和不确定条件下单流水线的成本目标函数,探讨模型参数(车辆载重、可运输车辆、安装时间窗宽度)与成本的关系,对装配式建筑调度管理提供有效建议。在装配式建筑多主体、多阶段的特点下,装配式建筑生产-运输集成调度优化根据构件安装时间窗统筹协调装配式建筑生产和运输阶段的调度方案,为装配式建筑调度问题提供科学合理的调度策略,进一步提升装配式建筑的建造效率。更多还原

【Abstract】 With rising labor costs and increasingly prominent resource and environmental issues,the construction industry urgently needs to transform to sustainable development and industrialized buildings.Prefabricated buildings can shorten construction time,reduce energy consumption,improve construction quality,and reduce pollution and labor.Although prefabricated buildings have many advantages,the development of prefabricated buildings has not met expectations,mainly because there are problems such as complex scheduling,insufficient integrated management level,and frequent uncertain events in the construction process of prefabricated buildings.Therefore,from the perspective of prefabricated building component suppliers,this thesis establishes a prefabricated building production-transportation integrated scheduling model based on installation time windows under certain and uncertain conditions.This thesis aims to improve the efficiency of prefabricated building resource allocation,and Improve the ability to deal with uncertain events during the construction of prefabricated buildings.This thesis studies the integrated scheduling optimization problem of prefabricated building production-transportation under certain and uncertain conditions.Under certain conditions,the objective function of minimizing the total cost is established.In the production stage,the traditional permutation flow shop scheduling model is used,and factors such as process interruption,parallelism,and non-working hours are considered.In the transportation stage,considering factors such as large components,limited installation space,and the load of transportation vehicles is much smaller than the order quantity,the transportation process is modeled as JIT on-time delivery and batch delivery.A segmented chromosome based on component production sequence and transportation sequence is designed.Through adaptive greedy tabu genetic algorithm(referred to as AGTGA)in the adaptive crossover and mutation,destruction,taboo,reconstruction operators to avoid falling into the local optimal solution of the genetic algorithm,better balance the local and global Search to reduce the number of invalid searches.And through the Taguchi experiment to determine the optimal combination of AGTGA parameter values,using27 examples to test the effectiveness and superiority of AGTGA.In order to avoid insufficient productivity of a single assembly line and further enrich the application scenarios of the algorithm,the production mode of single supplier and multiple assembly lines under uncertain conditions is studied.Triangular fuzzy numbers are used to represent the uncertainty of production processing time and transportation time,and trapezoidal fuzzy numbers are used to represent the uncertainty of installation time window.And based on the hierarchical relationship between multiple assembly line decisions,a two-level optimization model of order assembly line allocation and single assembly line production-transportation integrated scheduling is established,and the two-level optimization model based on AGTGA is used to solve it.In the upper-level optimization,the assembly lines are assigned to component orders,the order assembly lines allocation scheme with the lowest total cost of multiple pipelines is calculated.In the lower-level optimization,the fuzzy pessimistic satisfaction degree considering the risk preference of the decision maker and the anti-risk ability is introduced,and the component production and transportation sequence scheme with the largest fuzzy pessimistic satisfaction degree under the single assembly line is calculated.The effectiveness of AGTGA-based twolevel optimization solution for prefabricated building production-transportation integrated scheduling under uncertain conditions is tested by using a case.The cost of a single assembly line under certain and uncertain conditions is further analyzed,and the relationship between model parameters(vehicle load,transportable vehicles,and installation time window width)and cost is discussed,and effective suggestions are provided for the scheduling management of prefabricated buildings.Under the multi-subject and multi-stage characteristics of prefabricated buildings,the production-transportation integrated scheduling optimization of prefabricated buildings coordinates the scheduling scheme of the production and transportation stages of prefabricated buildings according to the component installation time window,and provides a scientific and reasonable solution for the prefabricated building scheduling problem.Scheduling strategy to further improve the construction efficiency of prefabricated buildings.更多还原