Spatial arrangement using deep reinforcement learning to minimise rearrangement in ship block stockyards

As the shipbuilding industry is an engineering-to-order industry, different types of products are manufactured according to customer requests, and each product goes through different processes and workshops. During the shipbuilding process, if the product is not able to go directly to the subsequent process due to physical constraints of workshop, it temporarily waits in a stockyard. Since the waiting process involves unpredictable circumstances, plans regarding time and space cannot be established in advance. Therefore, unnecessary movement often occurs when ship blocks enter or depart from the stockyard. In this study, a reinforcement learning approach was proposed to minimise rearrangement in such circumstances. For this purpose, an environment in which blocks are arranged and rearranged was defined. Rewards based on the simplified rules were logically defined, and simulation was performed for quantitative evaluation using the proposed reinforcement learning algorithm. This algorithm was verified using an example model derived from actual data from a shipyard. The method proposed in this study can be used not only to the arrangement problem of ship block stockyards but also to the various arrangement and allocation problems or logistics problems in the manufacturing industry.     

Block spatial scheduling modelling and application in shipbuilding

With the wide application of module-shipbuilding technology, problems related to block spatial scheduling occur in various working areas, and this restricts the productivity of shipbuilding. To address the problems and to obtain the optimum block sequence and spatial layout, typical block features and work plates were investigated. A heuristic spatial scheduling model was established based on the investigation and proposed strategies with the objective to minimise makespan. With the heuristic algorithm, a block spatial scheduling system was developed and implemented with real data from a large ship. Through the spatial scheduling system, visual results of daily block layouts and progress charts for all blocks can be easily obtained and work orders can also be created for site workers. Several other spatial scheduling methods are described and compared with the above-mentioned heuristic algorithm. The result shows that the heuristic algorithm is better than Cplex and a genetic algorithm in solving large-scale block scheduling, and the heuristic algorithm is better than a grid algorithm and manual scheduling in all aspects such as makespan, utilisation of work plates, runtime of scheduling and on-time delivery. The developed block spatial scheduling system is applied in a block production shop of a modern shipyard and shows good performance.     

Prediction of transverse and angular distortions of gas tungsten arc bead-on-plate welding using artificial neural network

The angular distortion and transverse shrinkage are often generated in gas tungsten arc (GTA) bead-on-plate welding process, which leads to additional costs of rework. Therefore, it is beneficial to estimate the welding deformations prior to bead-on-plate welding in terms of several process parameters. This paper presents the development of a back propagation neural (BPN) network model for the prediction of angular distortion and transverse shrinkage generated in GTA bead-on-plate welding process. The model is based on the results from finite element (FE) simulations. The GTA bead-on-plate welding for S304L stainless steel was simulated using finite element method, and experiments were conducted to verify the accuracy of the FE model. The experimental results were also used as testing samples for the BPN model. Welding speed, current and voltage were considered as the input parameters and the angular distortion and transverse shrinkage were the output parameters in the development of the BPN model. The correlation coefficients and percentage errors for all the samples were calculated to evaluate the prediction accuracy of BPN model. The results show that the BPN model developed in this study can predict the angular distortion and transverse shrinkage with reasonable accuracy.     

Theoretical prediction of welding distortion in large and complex structures

Welding technology is widely used to assemble large thin plate structures such as ships, automobiles, and passenger trains because of its high productivity. However, it is impossible to avoid welding-induced distortion during the assembly process. Welding distortion not only reduces the fabrication accuracy of a weldment, but also decreases the productivity due to correction work. If welding distortion can be predicted using a practical method beforehand, the prediction will be useful for taking appropriate measures to control the dimensional accuracy to an acceptable limit. In this study, a two-step computational approach, which is a combination of a thermoelastic-plastic finite element method (FEM) and an elastic finite element with consideration for large deformation, is developed to estimate welding distortion for large and complex welded structures. Welding distortions in several representative large complex structures, which are often used in shipbuilding, are simulated using the proposed method. By comparing the predictions and the measurements, the effectiveness of the two-step computational approach is verified.     

Theoretical prediction of welding distortion in large and complex structures

Welding technology is widely used to assemble large thin plate structures such as ships, automobiles, and passenger trains because of its high productivity. However, it is impossible to avoid welding-induced distortion during the assembly process. Welding distortion not only reduces the fabrication accuracy of a weldment, but also decreases the productivity due to correction work. If welding distortion can be predicted using a practical method beforehand, the prediction will be useful for taking appropriate measures to control the dimensional accuracy to an acceptable limit. In this study, a two-step computational approach, which is a combination of a thermo-elastic-plastic finite element method (FEM) and an elastic finite element with consideration for large deformation, is developed to estimate welding distortion for large and complex welded structures. Welding distortions in several representative large complex structures, which are often used in shipbuilding, are simulated using the proposed method. By comparing the predictions and the measurements, the effectiveness of the two-step computational approach is verified.     

低成本船用EH36高强钢板的研制及其性能

不添加价格昂贵的铌、钒元素,仅在C-Mn钢成分体系上添加钛元素细化晶粒,采用热机械控制工艺研制出低成本的船用EH36高强钢板,并对其力学性能、显微组织等进行了分析.结果表明:该船用EH36高强钢板成本较原来低,其性能满足标准要求,且强度和韧性均有较大的富余量;经人工应变时效后,其低温冲击韧性仍较好;经两种热输入焊接后,...     

低代码可视化AI技术加速船舶焊接工艺智能化

在焊接质量分析及检测的智能化应用研究创新过程中,需要利用工业机理模型和分析方法进行可视化、低代码化和智能化构建。焊接工艺低代码可视化分析提供了一种基于机理模型的信号分析方法,可以帮助船舶制造行业解决焊接过程中的质量问题。该软件包括U-I相图、特征分析和短路专属分析等分析功能,以船舶发动机焊接为案例进行实证分析,结合U-I相图等由用户自主构建焊缝质量评分体系,将其封装为AI模型实现实时质量评分,验证了该软件对焊接质量分析和评估的有效性,提高了焊接质量评价效率和准确性。     

船舶焊接工艺评定专家系统的研究与开发

焊接工艺评定是控制船舶焊接结构质量不可缺少的重要环节.本系统引入知识管理和专家推理,建立了基于网络环境的焊接工艺评定专家管理系统,实现了船舶焊接工艺评定过程中的数据和流程的全过程管理,对提升船舶焊接工艺评定的技术水平和推动企业的信息化建设都有着十分重要的意义.本文基于船舶焊接工艺评定专家管理系统的开发,论述了系统的总体...     

船舶焊接工艺知识管理技术研究

船舶焊接工艺设计涉及的工艺知识范围广泛,用到的信息和知识繁杂。因此,对船舶焊接工艺知识进行系统化表示,建立丰富的工艺知识库,不仅是船舶焊接工艺设计的基础,而且对船舶制造技术的提高具有重要意义。本文基于船舶焊接工艺知识管理系统（SWPKM）的开发,采用了面向对象建模和产生式规则技术,对船舶焊接工艺知识进行了全面分析,论述了船舶焊接工艺知识表示、工艺知识库构建等。     

CO2角焊中手工与角焊机作业方式的对比研究

利用工作研究的技术和方法对船厂焊接操作中常用的两种CO2角焊工作方式进行了研究,建立了标准化的操作程序,并为两种工作方式建立了评价指标及评价函数,从而得出适用于两种作业方式的选择条件,为合理选择手工焊作业与角焊机作业,提高焊接作业生产效率提供了依据。     

Spatial scheduling for shape-changing mega-blocks in a shipbuilding company

To overcome space restriction and to increase productivity, some shipbuilding companies use floating-docks on the sea instead of dry-docks on the land. In that case, a floating-crane that is capable of lifting very heavy objects (up to 3600?tons) is used to handle the blocks which are the basic units in shipbuilding processes, and therefore, very large blocks (also called mega-blocks) can be used to build a ship, but because there are some positional restrictions under which the mega-block assembly yard can be constructed, the space is the scarcest resource in the process. The focus of the research reported in this paper is to develop an efficient spatial schedule for the mega-block assembly yard. First, we develop a length-time two-dimensional packing model for this problem. Since the optimisation model cannot be solved using an analytical method, we propose a GA-based heuristic algorithm using computational geometry theory. Through performing a series of computational experiments, we finally show that the proposed spatial scheduling algorithm can provide solutions of good quality very efficiently.     

船用高强钢焊接技术的研究现状与展望

船用高强钢焊接技术是当下造船行业中的重要技术之一,高效、优质的船舶焊接技术已成为船舶工业制造产业长期战略的关键因素。从船用高强钢传统焊接技术、高效电弧焊接技术、高效新型焊接技术等方面介绍了船用高强钢焊接技术的研究现状,重点讨论了船用高强钢焊接技术的细分领域现状,分析了船用高强钢焊接技术的发展趋势。研究表明,船用钢板的板厚朝着厚板、大厚板方向发展,船用钢板的强度朝着高强度、超高强度方向发展,国内外船用高强钢焊接技术相差不大,高效混合焊接技术、高新焊接技术是未来船用高强钢焊接技术的研究热点。     

船体分段模块装配工艺决策研究

针对船体装配工艺设计自动化程度低，影响船舶建造周期和建造质量等问题，设计了分段模块的装配单元决策系统，建立了分段模块装配信息模型，基于模糊数学理论，应用模糊聚类方法实现了分段模块装配工艺的自动编制．实例表明，该方法是合理的和可行的．应用该方法在较高的装配层次上实现了船体模块装配工艺的制定，为船舶模块化生产提供了有效的工具．     

Minimising makespan on parallel machines with precedence constraints and machine eligibility restrictions

The block erection problem is defined as a parallel machine scheduling problem with precedence constraints and machine eligibility restrictions. A heuristic algorithm combined of the largest total amount of processing first rule (LTAP) and the enhanced smallest machine load first rule (ESML) is proposed to minimise makespan for the block erection in a shipyard. Finally, four lower bounds and the percentage of the reduced makespan compared with the current solutions are defined to evaluate the performance of the proposed algorithm. The experiments are performed on data selected from a shipbuilding company, and the results demonstrate that the presented algorithm can effectively find a good solution to minimise the makespan of the block erection problem.     

Optimisation of critical chain sequencing based on activities’ information flow interactions

One critique for the classic critical chain sequencing methods is that only resource constraints and logical relationships between activities are considered, while interactions of information flows are ignored. However, information flow interactions exist between almost all project activities and bring about big rework risk and rework costs in project management, especially for the long-term projects tied up with big amount of money. In this paper, we propose a new approach to sequence a project’s critical chain by considering activities’ information flow. Specifically, we use the design structure matrix to measure directions and intensities of information flows between activities. The criterion for critical chain sequencing is then to minimise the total coordination cost while considering the influence of feedbacks. The simulation results show that, as opposed to the traditional critical chain sequencing methods, our approach shortens average rework time and reduces the cost.     

日本两公司开发的大热输入焊接用高强度钢

为了提高船舶及建筑结构等的焊接效率,必须采用大热输入焊接等措施。介绍了日本各钢铁公司研究开发的多种新技术,包括利用TiN对热影响区晶粒长大进行有效控制,依靠铁素体形核核心手段实现晶粒细化,并利用控制热影响区微观组织以及采用TMCP控制轧制技术等。在综合利用这些技术的基础上,开发成功了各种适合于大热输入焊接的高强度钢,以确保大热输入条件下热影响区具有高的韧性,用于造船、建筑结构、桥梁等的建造。     

Dynamic lot-sizing with rework of defective items and minimum lot-size constraints

In most production processes, defective items may result from an imperfect production system and the need of reworking them is inevitable in many production environments. Despite the great importance of rework in real-world manufacturing, the body of literature is very limited. This paper deals with the effects of defective items and rework on the Capacitated Lot-Sizing Problem (CLSP). We present a mixed-integer programming formulation of the CLSP with rework of defective items and minimum lot-size constraints on production lots. The formulation describes an imperfect production process that leads to a fraction of defective items that have to be reworked before they can be sold to customers. Detailed numerical experiments show that while the occurrence of defective items significantly increases the computational times, reasonably sized minimum lot-size constraints, besides their practical importance, can be a good strategy to accelerate the solution process.     

Improving automotive dimensional quality by using principal component analysis

Dimensional quality is a measure of conformance of the actual geometry of products with the designed geometry. In the automotive body assembly process, maintaining good dimensional quality is very difficult and critical to the product. In this paper, a dimensional quality analysis and diagnostic tool is developed based on principal component analysis (PCA). In quality analysis, the quality loss due to dimensional variation can be partitioned into a mean deviation and piece-to-piece variation. By using PCA, the piece-to-piece variation can be further decomposed into a set of independent geometrical variation modes. The features of these major variation modes help in identifying the underlying causes of dimensional variation in order to reduce the variation. The variation mode chart developed in this paper provides the explicit and exact geometrical interpretation of variation modes, making PCA easily understood. A case study using an automotive body assembly dimensional quality analysis will illustrate the value and power of this methodology in solving actual engineering problems in a practical manner.     

Optimal locations of on-line and off-line rework stations in a serial production system

The production rate and product quality are two vital concerns for any manufacturing industry. Number of defective items reduces production rate and increases unit production cost. Moreover, if nonconforming items reach to the customers then manufacturer’s goodwill may drastically go down. Thus, quality inspection is treated as an inherent part of manufacturing. In this research, an N-stage serial production line with an inspection station at the end of it is considered to make decisions concerning this issue. On detecting a defective item at the end of the line it is scrapped or repaired at regular workstation or is sent to an off-line rework station for repair. Assuming each workstation produces a single type of defect a unit cost function is developed for alternative decisions on each type of defect. In order to minimise the unit cost of production and determine an appropriate decision for individual defect types, a fractional mixed integer nonlinear programming is formulated. After transformation to a mixed integer linear programming problem it is solved optimally. A small problem from garments industry is described in detail to show the solution procedure with a branch and bound method. Empirical tests with up to 40 workstations are permed to show the efficiency of the solution process.     

From value stream mapping toward a lean/sigma continuous improvement process: an industrial case study

Though lean manufacturing has been widely recognised for its effectiveness in continuously improving productivity, product quality, and on-time delivery to customers, the cost for hiring a full-time lean manufacturing engineer has kept many small businesses from implementing lean in their facilities. This paper presents a case study of lean implementation at a small manufacturer in the United States. Starting with collecting process information, a current value-stream map was created that reflected the current operation status. A future value stream map was then proposed to serve as a guide for future lean activities. Next, hurdles that kept the company from moving towards future state were identified. The ‘5 whys’ method was employed to reveal the root cause for each hurdle, followed by kaizen events proposed as solutions. In this case study, two kaizen events were proposed. For the first kaizen event, Taguchi experiment design was used to find the optimal machining parameters that reduced variation in a plasma cutting process. It consequently eliminated rework time and improved productivity. In the second kaizen event, implementation of rabbit chasing increased the system flexibility and consequently reduced inventory levels between work stations.