Mixed-model Production Lines Research Articles

AI for in-line vehicle sequence controlling: development and evaluation of an adaptive machine learning artifact to predict sequence deviations in a mixed-model production line

Customers in the manufacturing sector, especially in the automotive industry, have a high demand for individualized products at price levels comparable to traditional mass-production. The contrary objectives of providing a variety of products and operating at minimum costs have introduced a high degree of production planning and control mechanisms based on a stable order sequence for mixed-model assembly lines. A major threat to this development is sequence scrambling, triggered by both operational and product-related root causes. Despite the introduction of Just-in-time and fixed production times, the problem of sequence scrambling remains partially unresolved in the automotive industry. Negative downstream effects range from disruptions in the Just-in-sequence supply chain, to a discontinuation of the production process. A precise prediction of sequence deviations at an early stage allows the introduction of counteractions to stabilize the sequence before disorder emerges. While procedural causes are widely addressed in research, the work at hand requires a different perspective involving a product-related view. Built on unique data from a real-world global automotive manufacturer, a supervised classification model is trained and evaluated. This includes all the necessary steps to design, implement, and assess an AI-artifact, as well as data gathering, preprocessing, algorithm selection, and evaluation. To ensure long-term prediction stability, we include a continuous learning module to counter data drifts. We show that up to 50% of the major deviations can be predicted in advance. However, we do not consider any process-related information, such as machine conditions and shift plans, but solely focus on the exploitation of product features like body type, power train, color, and special equipment.

The effect of operational policies on production systems robustness: an aerospace case study

This study aims to evaluate the robustness of a low volume mixed model production line under different operating conditions. A real production line that builds six different aircraft heat exchangers is modelled, simulated and analysed under different operating conditions. A number of experiments are conducted in order to assess the effect of dispatching rules and disturbances related to reworks, and processing time variance on tardiness robustness. A penalty function to quantitatively assess tardiness is defined based on lead time at capacity and it is used to measure the robustness of the system. The results of the assessment are then discussed in order to give some practical guidance to production planners with controlling the line in the face of uncertainty or disturbances similar to those evaluated in the study.

Optimizing control dynamic complexity and production schedule

Over the last few decades sophisticated line balancing methodologies have been developed in response to the emerging trend of lean production in order to more effectively streamline production sequence scheduling. Newer approaches now demonstrate the potential to achieve lean production while at the same time simplifying what have become increasingly complex analytical methodologies and production sequences. Unlike past studies on line balancing for mixed-model production lines, in this paper the impacts of line balance on Control Dynamic Complexity are investigated and their deployment illustrated with a case study. Precedence matrices have been studied and a combined precedence matrix computed. The mathematical programming language Matlab is employed to illustrate state diagrams representing the relationships between the tasks and terminals. It is found that control dynamic complexities are higher for some particular sequences, and in such cases further explanations are offered. Furthermore, this study reveals that it is possible to minimize the complexities by better exploiting some aspects of line scheduling guidelines. Since line balancing and line scheduling are NP hard issues, this paper has also demonstrated ways in which computational analysis of multiple iterations can be significantly accelerated in future by deploying ant colony optimization techniques. Analysis of complexity level highlights what may otherwise be intuitive, that excessive sophistication of line balance techniques is very likely to be detrimental to an overall organization if resulting scheduling becomes difficult to practically monitor.

Evaluating manufacturing systems robustness: an aerospace case study

The scope of this work is the quantification of manufacturing systems robustness. Initially the concept of robustness is discussed and then a robustness definition & a metric are introduced. A framework that combines discrete event simulation and statistical analysis is also suggested for the evaluation of robustness. The efficacy of the metric and the framework are illustrated with a case study of a low–volume mixed-model production line that stems from aerospace industry. A set of scenarios are defined where the effect of different sources of disturbances on flowtime and tardiness robustness is quantified.

혼류생산 방식을 적용한 신개념 용접조립 기술 연구

Mixed-model production lines are often used in manufacturing systems. In production lines, different product types are simultaneously manufactured by processing small batches. This paper describes a new welding assembly technology involving the development of experimental models for a mixed-model production line in an automobile company. Due to the extensive number of models, the design of a welding assembly system is complicated. Performance evaluation is an important phase in the design of welding assembly lines in a mixed-model production environment. In this study, a new welding assembly technology for a mixed-model production method was used to weld the package tray and dash panel of a vehicle.

Modeling sequence scrambling and related phenomena in mixed-model production lines

In this paper we examine the various effects that workstations and rework loops with identical parallel processors and stochastic processing times have on the performance of a mixed-model production line. Of particular interest are issues related to sequence scrambling. In many production systems (especially those operating on just-in-time or in-line vehicle sequencing principles), the sequence of orders is selected carefully to optimize line efficiency while taking into account various line balancing and product spacing constraints. However, this sequence is often altered due to stochastic factors during production. This leads to significant economic consequences, due to either the degraded performance of the production line, or the added cost of restoring the sequence (via the use of systems such as mix banks or automated storage and retrieval systems). We develop analytical formulas to quantify both the extent of sequence scrambling caused by a station of the production line, and the effects of this scrambling on downstream performance. We also develop a detailed Markov chain model to analyze related issues regarding line stoppages and throughput. We demonstrate the usefulness of our methods on a range of illustrative numerical examples, and discuss the implications from a managerial point of view.

Multiobjective Fuzzy Mixed Assembly Line Sequencing Optimization Model

It can be deduced from previous studies that there exists a research gap in assembly line sequencing optimization model for mixed-model production lines. In particular, there is a lack of studies which focus on the integration between job shop and assembly lines using fuzzy techniques. Hence, this paper is aimed at addressing the multiobjective mixed-model assembly line sequencing problem by integrating job shop and assembly production lines for factories with modular layouts. The primary goal is to minimize the make-span, setup time, and cost simultaneously in mixed-model assembly lines. Such conflicting goals arise when switching between different products. A genetic algorithm (GA) approach is used to solve this problem, in which trapezoidal fuzzy numbers are implemented for variables such as operation and travelling time in order to generate results with higher accuracy and representative of real-case data.

SEQUENCING IN A NON-PERMUTATION FLOWSHOP WITH CONSTRAINED BUFFERS: APPLICABILITY OF GENETIC ALGORITH VERSUS CONSTRAINT LOGIC PROGRAMMING

Mixed model production lines consider more than one model being processed on the same production line in an arbitrary sequence. Nevertheless, the majority of publications in this area are limited to solutions which determine the job sequence before the jobs enter the line and maintains it without interchanging jobs until the end of the production line, which is known as permutation flowshop. This paper considers a nonpermutation flowshop. Resequencing is permitted where stations have access to intermediate or centralized resequencing buffers. The access to the buffers is restricted by the number of available buffer places and the physical size of the products. Two conceptually different approaches are presented in order to solve the problem. The first approach is a hybrid approach, using Constraint Logic Programming (CLP), whereas the second one is a Genetic Algorithm (GA). Improvements that come with the introduction of constrained resequencing buffers are highlighted. Characteristics such as the difference between the intermediate and the centralized case are analyzed, and the special case of semi dynamic demand is studied. Finally, recommendations are presented for the applicability of the hybrid approach, using CLP, versus the Genetic Algorithm.

Balancing and sequencing mixed-model U-lines with a co-evolutionary algorithm

A mixed-model production line is such a line where a variety of product models are produced. In U-lines used in the just-in-time production system, the strategy of mixing product models is often employed to provide various types of products to customers on time. Line balancing and model sequencing problems are important for an efficient use of mixed-model U-lines. Although the two problems are tightly interrelated with each other, prior research has considered them separately or sequentially. In this paper, a new approach using an artificial intelligence search technique, called co-evolutionary algorithm, is proposed to solve the two problems at the same time. To promote population diversity and search efficiency in the algorithm, we adopt strategies of localized evolution and steady-state reproduction, and develop methods of selecting environmental individuals and evaluating fitness. Efficient genetic representations and operator schemes are also provided. When designing the schemes, we take into account the features specific to the problems. The experimental results demonstrate that the proposed algorithm outperforms existing approaches.

Cyclic scheduling in synchronous production lines

In this paper we address the scheduling problem in unpaced synchronous mixed-model production lines operated under a cyclic scheduling policy. We first discuss operations of a production line with the synchronous transfer of parts. We then present an integer programming formulation of the problem. The problem, however, is W-hard, and for its exact solution we propose an implicit enumeration scheme. We discuss a property of the scheduling problem which allows us to effectively solve large size instances of the problem. We also present an approximate solution procedure with very good avenge performance. Useful managerial insights are obtained as we search for ways to improve the performance of synchronous lines. The relaxation of one of our original assumptions in the scheduling problem formulation results in an easy problem whose solution generates the absolute best in throughput performance configuration of the production line. Implementation of this solution, however, requires increasing the number of buffers in the line. We suggest other performance improvement ways to better balance the tradeoff between throughput and average Work-In-Progress (WIP) inventory in the line.

A parallel station heuristic for the mixed-model production line balancing problem

Demand for customized products and proliferation of optimal features have increased the need for flexible assembly systems that are capable of simultaneously producing multiple versions of similar products. Serial assembly systems have traditionally been used for the production of a single product type, and more recently for mixed model production. In this paper, a nonlinear integer program is proposed as a model for the production line balancing problem (PLBP). This problem entails the assignment of tasks to stages in a serial production line. The model allows mixed-model production and the use of identical parallel workstations at each stage of the serial production system. The objective function trades off idle workstation time with duplication of task-dependent equipment/ tooling cost. A heuristic is developed to create parallel workstations and assign tasks. Station utilization is also explicitly considered by using a threshold variable for target (acceptable) levels. The procedure is illustrated with an example. Computational experiments are also provided to analyse the performance of the heuristic.

A cost minimization procedure for mixed model production lines with normally distributed task times

This paper presents design procedures for mixed model production lines. The design objective is to minimize the total operating cost, composed of labor cost, in-process inventory cost, and set-up cost. This paper is an extension of our previous work in which we recognize the variability associated with task times and propose a procedure that can handle either deterministic or stochastic task times.

混合流れラインのシーケンシング

In order to operate effectively a mixed-model production line on which a variety of models of the same general product is intermixed, it is essential to determine the sequence in the flow of models. For a practical large-scale mixed-model production line which is subject to system variability, this paper develops a heuristic procedure which iteratively improves a given sequence into the optimum or near-optimum solution by inserting a model of the sequence into other positions and interchanging model pairs of the sequence. By extensively discussing solution times, the quality of solutions obtained by this procedure, and the probability that a given initial sequence reaches the optimum solution, the paper has shown that the procedure generates the optimum and near-optimum solutions highly effectively. Run times grow approximately as M2.2 with the size (M) of models, and linearly with the number of stations.