On solving the assembly line worker assignment and balancing problem via beam search

Certain types of manufacturing processes can be modelled by assembly line balancing problems. In this work we deal with a specific assembly line balancing problem that is known as the assembly line worker assignment and balancing problem (ALWABP). This problem appears in settings where tasks must be assigned to workers, and workers to work stations. Task processing times are worker specific, and workers might even be incompatible with certain tasks. The ALWABP was introduced to model assembly lines typical for sheltered work centers for the Disabled.     

A simulated annealing algorithm for multi-manned assembly line balancing problem

Assembly line balancing problems with multi-manned workstations usually occur in plants producing high volume products (e.g. automotive industry) in which the size of the product is reasonably large to utilize the multi-manned assembly line configuration. In these kinds of assembly lines, usually there are multi-manned workstations where a group of workers simultaneously performs different operations on the same individual product. However, owing to the high computational complexity, it is quite difficult to achieve an optimal solution to the balancing problem of multi-manned assembly lines with traditional optimization approaches. In this study, a simulated annealing heuristic is proposed for solving assembly line balancing problems with multi-manned workstations. The line efficiency, line length and the smoothness index are considered as the performance criteria. The proposed algorithm is illustrated with a numerical example problem, and its performance is tested on a set of test problems taken from literature. The performance of the proposed algorithm is compared to the existing approaches. Results show that the proposed algorithm performs well.     

2-ANTBAL: An ant colony optimisation algorithm for balancing two-sided assembly lines

Two-sided assembly lines are a special type of assembly lines in which workers perform assembly tasks in both sides of the line. This type of lines is of crucial importance, especially in the assembly of large-sized products, like automobiles, buses or trucks, in which some tasks must be performed at a specific side of the product. This paper presents an approach to address the two-sided mixed-model assembly line balancing problem. First, a mathematical programming model is presented to formally describe the problem. Then, an ant colony optimisation algorithm is proposed to solve the problem. In the proposed procedure two ants ‘work’ simultaneously, one at each side of the line, to build a balancing solution which verifies the precedence, zoning, capacity, side and synchronism constraints of the assembly process. The main goal is to minimise the number of workstations of the line, but additional goals are also envisaged. The proposed procedure is illustrated with a numerical example and results of a computational experience that exhibit its superior performance are presented.     

Quantitative postural load assessment for whole body manual tasks based on perceived discomfort

Many Korean workers are exposed to repetitive manual tasks or prolonged poor working postures that are closely related to back pain or symptoms of musculoskeletal disorders. Workers engage in tasks that require not only handling of heavy materials, but also assuming prolonged or repetitive non-neutral work postures. Poor work postures that have been frequently observed in the workplaces of shipbuilding shops, manufacturing plants, automobile assembly lines and farms often require prolonged squatting, repetitive arm raising and wrist flexion and simultaneous trunk flexion and lateral bending. In most manufacturing industries, workers have to assume improper work postures repetitively, several hundreds of times per day depending on daily production rate. A series of psychophysical laboratory experiments were conducted to evaluate the postural load at various joints. A postural load assessment system was then developed based on a macro-postural classification scheme. The classification scheme was constructed based on perceived discomfort for various joint motions as well as previous research outcomes. On the basis of the perceived discomfort, postural stress levels for the postures at individual joints were also defined by a ratio scale to the standing neutral posture. Laboratory experiments simulating automobile assembly tasks were carried out to investigate the relationship between body-joint and whole-body discomfort. Results showed a linear relationship between the two types of discomfort, with the shoulder and low back postures being the dominant factor in determining the whole body postural stresses. The proposed method was implemented into a computer software program in order to automate the procedure of analysing postural load and to enhance usability and practical applicability.     

Balancing and scheduling of flexible mixed model assembly lines

Mixed model assembly line literature involves two problems: balancing and model sequencing. The general tendency in current studies is to deal with these problems in different time frames. However, in today’s competitive market, the mixed model assembly line balancing problem has been turned into an operational problem. In this paper, we propose mixed integer programming (MIP) and constraint programming (CP) models which consider both balancing and model sequencing within the same formulation along with the optimal schedule of tasks at a station. Furthermore, we also compare the proposed exact models with decomposition schemes developed for solving different instances of varying sizes. This is the first paper in the literature which takes into account the network type precedence diagrams and limited buffer capacities between stations. Besides, it is the first study that CP method is applied to balancing and scheduling of mixed model assembly lines. Our empirical study shows that the CP approach outperforms the MIP approach as well as the decomposition schemes.     

Quantitative postural load assessment for whole body manual tasks based on perceived discomfort

Many Korean workers are exposed to repetitive manual tasks or prolonged poor working postures that are closely related to back pain or symptoms of musculoskeletal disorders. Workers engage in tasks that require not only handling of heavy materials, but also assuming prolonged or repetitive non-neutral work postures. Poor work postures that have been frequently observed in the workplaces of shipbuilding shops, manufacturing plants, automobile assembly lines and farms often require prolonged squatting, repetitive arm raising and wrist flexion and simultaneous trunk flexion and lateral bending. In most manufacturing industries, workers have to assume improper work postures repetitively, several hundreds of times per day depending on daily production rate. A series of psychophysical laboratory experiments were conducted to evaluate the postural load at various joints. A postural load assessment system was then developed based on a macro-postural classification scheme. The classification scheme was constructed based on perceived discomfort for various joint motions as well as previous research outcomes. On the basis of the perceived discomfort, postural stress levels for the postures at individual joints were also defined by a ratio scale to the standing neutral posture. Laboratory experiments simulating automobile assembly tasks were carried out to investigate the relationship between body-joint and whole-body discomfort. Results showed a linear relationship between the two types of discomfort, with the shoulder and low back postures being the dominant factor in determining the whole body postural stresses. The proposed method was implemented into a computer software program in order to automate the procedure of analysing postural load and to enhance usability and practical applicability.     

Upper Body and Finger Posture Evaluations at an Electric Iron Assembly Plant

This study evaluated the postures that workers used to perform 18 different tasks at an electric iron assembly plant by classifying the common simultaneous and individual postures of eight upper body segments and the fingers. The postures of the head, upper arms, lower arms, hand, and trunk were assigned to the categories of “Neutral,” “Bend,” “Twist,” and “Invisible.” The finger postures were also assigned to 14 categories. Overall, most workers bent the head, lower arms, and hands and used power grips to wrap all fingers around parts and tools. The upper arms and trunk were in neutral positions because the workers stood at a conveyor belt. Among 18 tasks, the task of “setting temperature” seemed the lightest work because most body segments were in neutral positions for more than 54.6% of a cycle time, while the task of “palletizing” seemed the most stressful work because all body segments were bent more than 54.4% of a cycle time. These posture differences among the tasks result from the different task characteristics of workplace design, task difficulty, and work duration. It would be useful in designing workplace and evaluating physical workload. © 2012 Wiley Periodicals, Inc.     

Model and heuristics for the Assembly Line Worker Integration and Balancing Problem

We propose the Assembly Line Worker Integration and Balancing Problem (ALWIBP), a new assembly line balancing problem arising in lines with conventional and disabled workers. The goal of this problem is to maintain high productivity levels by minimizing the number of workstations needed to reach a given output, while integrating in the assembly line a number of disabled workers. Being able to efficiently manage a heterogeneous workforce is especially important in the current social context where companies are urged to integrate workers with different profiles. In this paper we present mathematical models and heuristic methodologies that can help assembly line managers to cope with this additional complexity. We demonstrate by means of a robust benchmark how this integration can be done with losses of productivity that are much lower than expected.     

Applying incremental Deep Neural Networks-based posture recognition model for ergonomics risk assessment in construction

Monitoring and assessing awkward postures is a proactive approach for Musculoskeletal Disorders (MSDs) prevention in construction. Machine Learning models have shown promising results when used in recognition of workers’ posture from Wearable Sensors. However, there is a need to further investigate: i) how to enable Incremental Learning, where trained recognition models continuously learn new postures from incoming subjects while controlling the forgetting of learned postures; ii) the validity of ergonomics risk assessment with recognized postures. The research discussed in this paper seeks to address this need through an adaptive posture recognition model– the incremental Convolutional Long Short-Term Memory (CLN) model. The paper discusses the methodology used to develop and validate this model’s use as an effective Incremental Learning strategy. The evaluation was based on real construction workers’ natural postures during their daily tasks. The CLN model with “shallow” (up to two) convolutional layers achieved high recognition performance (Macro F1 Score) under personalized (0.87) and generalized (0.84) modeling. Generalized CLN model, with one convolutional layer, using the “Many-to-One” Incremental Learning scheme can potentially balance the performance of adaptation and controlling forgetting. Applying the ergonomics rules on recognized and ground truth postures yielded comparable risk assessment results. These findings support that the proposed incremental Deep Neural Networks model has a high potential for adaptive posture recognition. They can be deployed alongside ergonomics rules for effective MSDs risk assessment.     

An effective hybrid honey bee mating optimization algorithm for balancing mixed-model two-sided assembly lines

Mixed-model two-sided assembly lines are widely used in a range of industries for their abilities of increasing the flexibility to meet a high variety of customer demands. Balancing assembly lines is a vital design issue for industries. However, the mixed-model two-sided assembly line balancing (MTALB) problem is NP-hard and difficult to solve in a reasonable computational time. So it is necessary for researchers to find some efficient approaches to address this problem. Honey bee mating optimization (HBMO) algorithm is a population-based algorithm inspired by the mating process in the real colony and has been applied to solve many combinatorial optimization problems successfully. In this paper, a hybrid HBMO algorithm is presented to solve the MTALB problem with the objective of minimizing the number of mated-stations and total number of stations for a given cycle time. Compared with the conventional HBMO algorithm, the proposed algorithm employs the simulated annealing (SA) algorithm with three different neighborhood structures as workers to improve broods, which could achieve a good balance between intensification and diversification during the search. In addition, a new encoding and decoding scheme, including the adjustment of the final mated-station, is devised to fit the MTALB problem. The proposed algorithm is tested on several sets of instances and compared with Mixed Integer Programming (MIP) and SA. The superior results of these instances validate the effectiveness of the proposed algorithm.     

基于优化DRG的三维人体点云骨架提取方法

针对离散Reeb图（Discrete Reeb Graph,DRG）描述人体骨架时分支部位骨架线偏离中轴的问题,采用了能量函数最小化的方法对DRG曲线进行优化。将人体模型的DRG曲线作为初始骨架,定义其能量函数,在点云模型的距离场梯度的作用下,迭代地调整偏离中轴目标段的曲线位置使其逐渐逼近中轴,能量函数最小时得到优化的骨架。将该算法应用于同一模特四个不同姿势和四个不同模特同一姿势的人体点云模型,并与基于拉普拉斯算子的点云收缩的骨架提取方法进行了比较。结果表明,该算法能够很好地适应各种不同姿势和体型,模型分叉部位的特征得到更加完善的描述,得到的骨架曲线更接近模型的中轴。     

Convolutional long short-term memory model for recognizing construction workers’ postures from wearable inertial measurement units

This paper proposes using Deep Neural Networks (DNN) models for recognizing construction workers’ postures from motion data captured by wearable Inertial Measurement Units (IMUs) sensors. The recognized awkward postures can be linked to known risks of Musculoskeletal Disorders among workers. Applying conventional Machine Learning (ML)-based models has shown promising results in recognizing workers’ postures. ML models are limited – they reply on heuristic feature engineering when constructing discriminative features for characterizing postures. This makes further improving the model performance regarding recognition accuracy challenging. In this paper, the authors investigate the feasibility of addressing this problem using a DNN model that, through integrating Convolutional Neural Networks (CNN) with Long Short-Term Memory (LSTM) layers, automates feature engineering and sequential pattern detection. The model’s recognition performance was evaluated using datasets collected from four workers on construction sites. The DNN model integrating one convolutional and two LSTM layers resulted in the best performance (measured by F1 Score). The proposed model outperformed baseline CNN and LSTM models suggesting that it leveraged the advantages of the two baseline models for effective feature learning. It improved benchmark ML models’ recognition performance by an average of 11% under personalized modelling. The recognition performance was also improved by 3% when the proposed model was applied to 8 types of postures across three subjects. These results support that the proposed DNN model has a high potential in addressing challenges for improving the recognition performance that was observed when using ML models.     

Assembly line balancing and group working: A heuristic procedure for workers’ groups operating on the same product and workstation

In this paper, we examine an assembly line balancing problem that differs from the conventional one in the sense that there are multi-manned workstations, where workers’ groups simultaneously perform different assembly works on the same product and workstation. This situation requires that the product is of sufficient size, as for example in the automotive industry, so that the workers do not block each other during the assembly work. The proposed approach here results in shorter physical line length and production space utilization improvement, because the same number of workers can be allocated to fewer workstations. Moreover, the total effectiveness of the assembly line, in terms of idle time and production output rate, remains the same. A heuristic assembly line balancing procedure is thus developed and illustrated. Finally, experimental results of a real-life automobile assembly plant case and well-known problems from the literature indicate the effectiveness and applicability of the proposed approach in practice.     

Robotic milling posture adjustment under composite constraints: A weight-sequence identification and optimization strategy

Industrial robots are widely used for milling complex parts in restricted spaces owing to their multiple degrees of freedom and flexible postures. To plan posture trajectory for robot machining with high precision under multiple constraints, this study establishes composite constraint models with constraint boundary solutions. An improved gray relation analysis model is adopted to identify the weight-sequences among the composite constraints. The correlation degrees of the postures of the robot can be dynamically quantified between arbitrary cutter locations by applying weight sequence identification, which is conducive to fulfilling attractive orientations in artificial potential fields. In addition, this study proposes an initial posture determination strategy based on the optimization principle of minimizing the rotated energy in global postures. Consequently, an artificial potential planning model is applied to the implement posture adjustment of the robot end effector. During simulation and experimental validation, the proposed posture adjustment strategies with optimized initial postures and identified weight-sequences achieve a significant improvement in both the six-joint motion performance and machining precision quality in robotic milling.     

Mixed model assembly line design in a make-to-order environment

Mixed model assembly lines can be found today in many industrial environments. With the growing trend for greater product variability and shorter life cycles, they are replacing the traditional mass production assembly lines. In many cases, these lines follow a ‘make-to-order’ production policy, which reduces the customer lead-time, and is expressed in a random arrival sequence of different model types to the line. Additional common characteristics of such mixed model lines in a make-to-order environment are: small numbers of work stations, a lack of mechanical conveyance, and highly skilled workers. The design problem of mixed model assembly lines in a make-to-order environment is addressed in this paper. A mathematical formulation is presented which considers the differences between our model and traditional models. A heuristic that minimizes the number of stations for a predetermined cycle time is developed consisting of three stages: the balancing of a combined precedence diagram, balancing each model type separately subject to the constraints resulting from the first stage, and a neighborhood search based improvement procedure.     

Effects of vehicle interior geometry and anthropometric variables on automobile driving posture

The effects of vehicle package, seat, and anthropometric variables on posture were studied in a laboratory vehicle mockup. Participants (68 men and women) selected their preferred driving postures in 18 combinations of seat height, fore-aft steering wheel position, and seat cushion angle. Two seats differing in stiffness and seat back contour were used in testing. Driving postures were recorded using a sonic digitizer to measure the 3D locations of body landmarks. All test variables had significant independent effects on driving posture. Drivers were found to adapt to changes in the vehicle geometry primarily by changes in limb posture, whereas torso posture remained relatively constant. Stature accounts for most of the anthropometrically related variability in driving posture, and gender differences appear to be explained by body size variation. Large intersubject differences in torso posture, which are fairly stable across different seat and package conditions, are not closely related to standard anthropometric measures. The findings can be used to predict the effects of changes in vehicle and seat design on driving postures for populations with a wide range of anthropometric characteristics.     

The effect of upper-extremity posture on maximum grip strength

The objective of this study was to determine and document the position of peak grip strength in different shoulder, elbow, and wrist posture combinations. Fifteen (15) male subjects performed maximal exertions with their dominant hands in nine wrist postures, three elbow postures, and two shoulder postures. Analysis of the data revealed that shoulder and elbow angles had significant effect upon the grip strength. Similarly, it was seen that grip strength at elbow at 135 degrees flexion was significantly different from those with elbow at 90 and 180 degrees. Further, the results revealed that peak grip strength occurred at a combined posture of shoulder abducted 0 degrees, elbow flexed 135 degrees, and the wrist in the neutral posture. Decrements of up to 42% in grip strength could be seen as elbow and wrist angles deviated. This means that use of handtools at deviated postures of shoulder, elbow, and wrist would decrease the percent of MVC at which a worker operated. Hence, the implementation of the finding of this study might be a reduction in the risk of injury, increase in productivity, and well-being of the workers.     

Posture variation among office workers when using different information and communication technologies at work and away from work

Office workers perform tasks using different information and communication technologies (ICT) involving various postures. Adequate variation in postures and muscle activity is generally believed to protect against musculoskeletal complaints, but insufficient information exists regarding the effect on postural variation of using different ICT. Thus, this study among office workers aimed to determine and compare postures and postural variation associated with using distinct types of ICT. Upper arm, head and trunk postures of 24 office workers were measured with the Physiometer® over a whole day in their natural work and away-from-work environments. Postural variation was quantified using two indices: APDF_(90-10) and EVA_(sd). Various ICT had different postural means and variation. Paper-based tasks had more non-neutral, yet also more variable postures. Electronics-based tasks had more neutral postures, with less postural variability. Tasks simultaneously using paper- and electronics-based ICT had least neutral and least variable postures. Tasks without ICT usually had the most posture variability. Interspersing tasks involving different ICT could increase overall exposure variation among office workers and may thus contribute to musculoskeletal risk reduction.     

Postural discomfort and perceived exertion in standardized box-holding postures

To help in the design or redesign of workplaces it would be helpful to know in advance the postural stress consequences of a wide range of body postures. This experiment evaluated 168 postures chosen to represent those in the Ovako Working-posture Analysing System (OWAS) using Rated Perceived Exertion (RPE) and Body Part Discomfort (BPD) measures. The postures comprised all combinations of three arm postures, four back postures, seven leg postures and two forces (weights of held boxes). Twelve male subjects held each posture for a fixed duration (20 s) before providing RPE and BPD ratings. Analysis of the ratings gave highly significant main effects, with the major driver being the object weight. As each factor was varied, the largest effect was on the body region corresponding to that factor. A simple main-effects-only additive model explained 91% of the variance of RPE means for the postures.     

A beam search approach for solving type II robotic parallel assembly line balancing problem

In a robotic assembly line, a series of stations are arranged along a conveyor belt and a robot performs on tasks at each station. Parallel assembly lines can provide improving line balance, productivity and so on. Combining robotic and parallel assembly lines ensure increasing flexibility of system, capacity and decreasing breakdown sensitivity. Although aforementioned benefits, balancing of robotic parallel assembly lines is lacking – to the best knowledge of the authors- in the literature. Therefore, a mathematical model is proposed to define/solve the problem and also iterative beam search (IBS), best search method based on IBS (BIBS) and cutting BIBS (CBIBS) algorithms are presented to solve the large-size problem due to the complexity of the problem. The algorithm also tested on the generated benchmark problems for robotic parallel assembly line balancing problem. The superior performances of the proposed algorithms are verified by using a statistical test. The results show that the algorithms are very competitive and promising tool for further researches in the literature.