Automated design of multi-station assembly lines

This article presents a novel approach for the automated design of assembly lines that combines the assembly line balancing problem with resource selection and the positioning of the chosen resources into one single optimisation problem. Existing approaches for the automated planning of assembly plants either focus on one planning step or work through different planning steps sequentially. So far, no method exists that sufficiently takes into account the interdependency between the selection and positioning of resources. This article addresses this problem by presenting a bi-level optimisation approach for the automated design of assembly lines. A nested genetic algorithm is used to solve an assembly line balancing problem that includes the selection of production resources while simultaneously considering the layouting options for the chosen resources. Three examples for the evaluation and validation of the algorithm are presented. The presented approach is economically promising as the design of assembly lines requires a lot of expert knowledge and is still mostly done manually.

Assembly lines with fractional tasks increase the throughput with a better workload balancing among the stations. The possibility to share tasks between two consecutive stations can also support the design of assembly lines with redundancies of the tasks. The redundancy of the tasks allows reacting to short failures or variations in processing time. This paper proposes a framework to support the design and control of redundancy assembly lines. The first problem concerns a modified mixed-integer linear programming model used for the fractional allocation assembly line problem to design redundancy assembly lines. Then, this research proposes control policies to manage the allocation of the tasks shared between two consecutive stations. Finally, the simulation evaluates the performance of the proposed methods to handle short failures and uncertain processing times. The numerical results show the potential improvements of the proposed methods with a limited number of redundancies. The main improvements concern the reduction of throughput time and work in process.

Purpose – This study seeks to present a simple assembly line design and its balance for a low‐volume manufacturing company.Design/methodology/approach – The study presents experiences with the design and implementation of a simple assembly line. The implementation concerns three aspects; design and construction of the assembly line, the assembly analysis of the product, and then balancing of the line. It also discusses construction and implementation difficulties of this tactical tool in the case company.Findings – The study presents some outcomes from the design, implementation, and balancing of an assembly line for SMEs.Research limitations/implications – The study is limited by the case company and its experience.Practical implications – This study is not pure theoretical study, it has application stages for industry, and it provides some real interface for the people from SMEs.Originality/value – The approach has an original value in respect of implementation of assembly line for a small manufacturing...

Multi-objective Early-stage Design of Automotive Hybrid Assembly Lines

Assembly lines commonly run for dozens of years before being decommissioned. As product families may evolve several times per year by following the needs of sales and marketing, process engineers reconfigure the lines several dozens of times throughout their life cycle. If the line is not flexible enough, these reconfigurations may be costly, and they can lead to poor efficiency. The present work investigates the possibility of designing a line while accounting for product evolution throughout the life cycle of the line. The evolution of the product family is unknown and we consider a robust optimisation approach. We study a mixed-model assembly line, where each station contains a worker/robot and its equipment. The line produces different product models from the same family, and a reconfiguration occurs when a new product model replaces one of the current variants in the product family. Reconfiguration re-arranges resources and equipment pieces, and it can re-assign some tasks. In this study, we formulate a novel Mixed-Integer Linear Programming (MILP) that minimizes the total cost of the initial design and future reconfigurations of the line over some future product family evolution for the worst case. We consider the worst-case among different scenarios that represent possible production requirements of the new product model. An adversarial approach is also developed to solve large-size instances. We perform computational experiments on the benchmark data from the literature. The results show the proposed adversarial approach performs well, and the proposed robust model significantly reduces the design and reconfiguration costs when compared to the classical approach that designs and reconfigures by accounting only for the current product family.

AutomationML as Seamless Data Exchange Format for Integration of Automated Planning Tools for Assembly Line Design

The most important factor affecting efficiency and ergonomic risk levels in an assembly line design is the problem of assigning certain tasks to certain stations, namely the assembly line balancing problem. In the literature, assembly line balancing problem has often been studied, but studies that consider ergonomic risks are deficient. Recently, it has been one of the issues that have started to attract great attention with the realization of health problems caused by assembly lines. To this end, in this study, a bi‐objective mathematical model is developed that considers balancing assembly line station time and ergonomic risk levels, simultaneously. It is aimed to minimize both station time and the total deviations of ergonomic risk scores for the stations. Weighted sum and conic scalarization methods are applied to solve the bi‐objective model. To analyze the outcomes of the developed model, an application is proposed and tested on a real industrial case, at a home appliance assembly line. The deployment of the OMAX method is a contribution to the literature since it shows an analysis tool which evaluates the results of assembly line balancing. This method evaluates the performance of the stations based on different criteria such as station time and ergonomic risk. The number of high‐risk stations is obtained as 13 in the single‐objective model aiming to minimize the station time, while it is found to be nine in the bi‐objective model solved with CSM, without an increase in the total number of stations.

A Simulation Study on the Layout Design of Micro Assembly Line for Lens Module

Design of assembly lines is a knowledge intensive process relying significantly on experience and associated with high costs, long lead times and high probability of risks and reworks. This paper presents a methodology to support the early stage design of assembly lines through Knowledge Based Engineering (KBE). In this introductory paper the authors discuss the methodology to implement the KBE approach. A set of system engineering rules is extracted from direct interviews and domain best practices and knowledge about product and process is formalized. These rules will be implemented in a user-friendly platform allowing the design of first phase line layout by taking the defined system requirements (e.g. cycle time) as input. Then, the KBE approach is extended to a specific case study taken from the powertrain sector.

The planning of assembly lines has always been a complex task, as a large number of factors that influence and disrupt production must be taken into account. Uncertainties in the early planning stages necessitate adjustments to the initial assumptions for the detailed assembly line plan throughout the planning project. Notably, the experience of assembly planners enables the integration of changing requirements into the planning process. This experience-based methodology is reaching its limits due to the increasing complexity of product variants and product portfolios. Newer data-driven approaches in assembly planning enable high productivity and the targeted use of synergies in assembly lines by utilising descriptive systematics that facilitate digital processing, thereby creating the basis for utilising digital planning assistance systems. For example, assembly precedence matrices (APMs) can structure and digitally represent the dependencies between assembly operations.If product variants with partially different assembly contents—whether in terms of the quantity of assembly operations or the specific activities within assembly operations—are processed within an assembly line, the individual consideration of these variants is no longer sufficient to optimise the system. To enable holistic optimisation, the synchronisation of APMs ensures consistency across variants. This is achieved through an expansion step, which integrates all relevant assembly operations into every APM within the product portfolio, followed by a transformation through row and column rearrangements. The synchronisation thus creates a daily basis for describing the assembly precedence relationships (APRs) of different product variants within the same product family.The scientific findings gained in this paper result from combining synchronised APMs of a mixed-model assembly line (MMAL) into a joint APM. The result includes defining operators for combining unified APM grids for a given product portfolio with different cell entries, taking into account application-related aspects. By combining the APMs of the product portfolio under consideration, APRs for the assembly procedure analysis can be described holistically as part of the assembly planning process. This enables analyses and optimisations for planning, such as assembly line balancing (ALB), the design of hybrid assembly structures or the dimensioning of assembly system networks.

PurposeCycle time fluctuations in assembly lines are one of the important reasons of re‐balancing. As a result of re‐balancing of assembly lines, it will be necessary to change task sequences or equipment locations. The purpose of this paper is to find the task sequence which enables assembly line balancing (ALB) with minimum number of stations (NS) for different cycle times such that tasks and equipment or fixture locations remain unchanged.Design/methodology/approachIn this paper a heuristic which consist of two stages is proposed to find a common task sequence for different cycle times in assembly lines.FindingsIt is shown that optimal NS for different cycle times can be achieved with a fixed task sequence.Research limitations/implicationsThe approach is limited to a single model case. Model variety together with cycle time variety can be investigated in further studies.Practical implicationsAssembly lines which require less time and cost for re‐balancing can be easily designed by the proposed approach.Originality/valueALB problem is handled with a new viewpoint. Also, it is observed that the proposed approach serves as a bridge between assembly line design and balancing. In this regard, it is thought to have an important place in the ALB literature.

Clustering of geometric objects is a very familiar and important problem in many different areas of applications as well as in the theoretical foundation of some modern fields of computer science. This paper describes how design problems, especially the design of an assembly line, can be transformed into a clustering problem. In order to solve the problem for large sizes of input data we introduce a structure, called Voronoi Tree, which applied to our real world data (assembly line design) did not only reduce the time to get a feasible design of an assembly line dramatically, but additionally increased the value of the design by more than 30% (in comparison with standard design methods). In addition to this we introduce a clustering method which is of interest for those applications which can be transformed to planar clustering problems. In this particular case it is possible to compute an (hierarchically) optimized clustering with resp. to a large class of clustering measures in timeO(nn1/2log3n+UF(n)nn1/2+PF(n)) [n: number of points;UF(n), PF(n) dependent on the chosen clustering measure].

Traditionally assembly line balancing problems have been examined under the assumption of serial work station design. Thus no pair of work stations could be simultaneously working on the same copy of a product. This paper presents two models for which this assumption is dropped. The algorithms developed for operational design of such assembly lines are presented along with sample problems to demonstrate the computational procedures and advantages of these models.

The total cost of assembly is a critical factor in robotic assembly line balancing, as it encompasses all the costs associated with the assembly line, including initial costs, setup, maintenance, and energy cost. This study introduces a different approach to the robotic assembly line balancing problem, with a dual focus on minimizing both cycle time and overall assembly costs. The effectiveness of the proposed approach is validated through three case study problems taken from the literature and results are compared to traditional assembly allocation methods. For case study 1, 89.4% (42 out of 47) of the solutions achieved a lower total cost, and 34% (16 out of 47) of the solutions utilized fewer workstations; and for case study 2, 96.4% (108 out of 112) of the solutions achieved a lower total cost, and 58.9% (66 out of 112) of the solutions utilized fewer workstations for the same cycle time. These results demonstrate a significant savings in cost and a notable improvement in workstation efficiency for a substantial portion of the solutions. This comprehensive approach allows an effective resource allocation, reduces inefficiencies, and enhances the overall cost-effectiveness and performance of the robotic assembly line. It also supports decision-makers in selecting more sustainable and economically viable assembly line solutions that optimize both productivity and energy efficiency.

Highly automated manufacturing systems offer several advantages and are widely introduced as a strategy to improve the performance of manufacturing organizations. Various forms of automated equipment, such as industrial robots or flexible automatic machines, are used extensively in high-volume industrial production. Especially for flow production systems automation has advanced considerably. This type of production system can, for instance, be found in the automotive, home appliance, or electronics industry where highly automated flow lines are mainly implemented for safety, quality, and productivity reasons. A significant challenge in the operation of highly automated assembly lines is the occurrence of equipment failures which impair the throughput rate. Therefore, buffer space is allocated between the stations of an assembly line in order to achieve a desired throughput rate in spite of equipment failures. However, the installation of buffer space requires considerable investments and also leads to an increase of the average work-in-process inventory in the line. A different approach to achieve a desired throughput rate despite equipment failures is a redundant configuration, in which downstream stations automatically take over the operations of failed stations in the event of failure. The throughput loss in these situations mainly depends on the level of redundancy designed into the system. We present an assembly line balancing model for automated assembly lines which maximizes the lines’ level of redundancy for a given number of stations. In a numerical analysis we demonstrate the effectiveness of the approach and show that a redundant configuration allows for significant reductions of required buffer sizes in flow lines with unreliable equipment for a given throughput rate.