Palletizing Task Research Articles

Biomechanical differences in experts' and novices' footstep patterns during a palletizing task

Very few studies have examined differences between experts' and novices' foot positioning and movements during manual materials handling tasks. The impact of footstep patterns on low back loading needs to be better understood. The goals of this study were to characterize foot placement and movements in novices and experts and to assess their impact on back loading considering the height of grasp. The task consisted in transferring 24 15 kg boxes from a pallet to another. Foot placement and movements were classified with a recently developed taxonomy. Results show that experts' feet remained static more often than novices' feet during the lifting phase. Positioning the feet towards the deposit site during lifting increased asymmetrical moments, especially for novices. Positioning one foot forward increased asymmetrical moments for novices. Overall, footstep strategies are an effective indicator of low back exposure and should be considered in ergonomic studies.

Adjustments in Shoulder and Back Kinematics during Repetitive Palletizing Tasks.

Repetitive task performance is a leading cause of musculoskeletal injuries among order-picking workers in warehouses. The repetition of lifting tasks increases the risk of back and shoulder injuries among these workers. While lifting in this industry is composed of loaded and unloaded picking and placing, the existing literature does not address the separate analysis of the biomechanics of the back and shoulder for these events. To that end, we investigated the kinematics of the back and shoulder movements of nine healthy male participants who performed three sessions of a simulated de/palletization task. Their back and shoulder kinematics were sensed using an optical motion capture system to determine the back inclination and shoulder flexion. Comparison of the kinematics between the first and last sessions indicated statistically significant changes in the timings, angles, coordination between the back and shoulder, and moments around the shoulder (). The majority of the significant changes were observed during the loaded events, which confirms the importance of the separation of these events for biomechanical analysis. This finding suggests that focusing worker evaluation on the loaded periods can provide important information to detect kinematic changes that may affect musculoskeletal injury risk.

Real-time Planning Robotic Palletizing Tasks using Reusable Roadmaps

This paper focuses on robotic motion planning for performing the palletizing or de-palletizing tasks. In such tasks, a robot usually iterates similar pick-and-place for several times. Considering such feature of the tasks, we propose two motion planning approaches named reusable Probabilistic Roadmap Method (PRM) and reusable Rapidly-exploring Random Tree Star (RRT*) where both methods utilize the previously constructed roadmaps in the conventional PRM and RRT*, respectively. We experimentally confirm that both methods significantly save the calculation time needed for motion planning compared to the conventional planning methods.

Real-time Planning Robotic Palletizing Tasks using Reusable Roadmaps

Real-time Planning Robotic Palletizing Tasks using Reusable Roadmaps

Increasing ABB FlexPicker Robot’s Degree of Freedom (DOF) using Flexible End Effector

The commissioning of a high speed parallel robot such as the ABB Flexpicker Robot into the industrial environment provides greater flexibility in picking, packaging or palletizing tasks. However, current configuration of this robot only allows up to four-degree-of-freedom (DOF) which restricted its function to perform simple or complex automated assembly task. Therefore, the possibility of increasing one additional degree-of-freedom for the parallel robot was investigated. By attaching a flexible vacuum hose to the robot’s end effector, it can be controlled by using a servomotor that can manipulate its angle of suction up to 90 degrees. With this new configuration, one DOF has been introduced for this robot. Several tests had been conducted to evaluate the flexibility and performance of the end effector which includes varying loads and angles. From the analysis, it is found that the end-effector can lift weights up to 1kg due to sufficient vacuum pressure; however, it could perform bending up to 90 degrees for maximum load of 600 grams only.

Sex differences in lifting strategies during a repetitive palletizing task

Forty-five manual material handlers (15 females, 15 expert males and 15 novice males) performed series of box transfers under conditions similar to those of large distribution centers. The objective of the study was to verify whether sex differences in joint motions and in back loading variables (L5/S1 moments) exist during multiple box transfers. The task consisted in transferring 24 15-kg boxes from one pallet to another (4 layers of boxes; 6 boxes/layer: 3 in the front row, 3 in the back) at a self-determined pace and then at an imposed pace of 9 lifts/min. Full-body 3D kinematic data were collected as well as external foot forces. A dynamic 3D linked segment model was used to estimate the net moments at L5/S1. The results show that the peak L5/S1 moment during lifting for females was significantly lower than for males, but once normalized to body size the difference disappeared. In general, the female workers were very close to the posture adopted by the novice males at the instant of the peak resultant moment. However, females were closer to the box than the male workers. One major sex difference was seen when lifting from the ground, with the use of interjoint coordination analyses. Female workers showed a sequential motion initiated by the knees, followed by the hip and the back, while expert males showed a more synchronized motion. The lifting strategy of females likely stretches lumbar spine passive tissues, which in turn put them at greater risk of back injuries. As observed in our previous studies, these differences between expert males, novice males and females are especially notable when the box is lifted from the ground.

The effects of transfer distance on spine kinematics when placing boxes at different heights

Twisting and lateral bending motions in repetitive lifting tasks are associated with occupational low back injuries and can be challenging to reduce with engineering controls. This study tested the hypothesis that twisting and lateral bending can be reduced by changing the transfer distance. Eighteen males, with no material handling experience lifted 10.9 kg boxes from 0.9 m above the floor and placed the boxes at a destination located 0.50, 0.75, 1.00, 1.25, 1.50, or 1.75 m away and at heights of 0.5 m, 0.9 m, and 1.3 m above the floor. Overall, twisting and forward bending decreased with increased transfer distance when placing the box. Conversely, the lateral bending when lifting and placing the box increased with increasing transfer distance. In short, having a transfer distance between 1 and 1.25 m when performing palletizing tasks to different heights may optimally balance spine kinematics, back injury risk, and productivity measures.

Lifting strategies of expert and novice workers during a repetitive palletizing task

Thirty manual material handlers (15 experts and 15 novices) were invited to perform series of box transfers under conditions similar to those of large distribution centers. The objective of the present study was to verify whether multiple box transfers leading to fatigue would also lead to differences between expert and novice workers in joint motions and in back loading variables (L5/S1 moments). The task consisted in transferring 24 15-kg boxes from one pallet to another (4 layers of boxes; 6 boxes/layer: 3 in the front row, 3 in the back) at a self-determined pace and then at an imposed pace of 9 lifts/min for a total of 240 lifts. The underlying idea was to set a challenging task that would force the experts to use their skills. Full-body 3D kinematic data were collected as well as external foot forces. A dynamic 3D linked segment model was used to estimate the net moments at L5/S1. The results clearly show that the experts bent their lumbar spine less (10° less) and were closer (4 cm) to the box than novice workers. Knee flexions were similar in both groups except when the box was lifted from ground level (expert ≈ 71°, novice ≈ 48°). The peak resultant moment was not statistically different (expert = 168 Nm, novice = 184 Nm) although experts had lower values on average than novices when lifting heights (and deposit heights) of the boxes increased. Therefore, experts differed from novice workers mostly in the posture-related variables. These differences are especially important to consider when the box is located on the ground, as the back posture and back loading are then at their greatest magnitude and could have a major impact on the distribution of internal forces on the spine.

Biomechanical evaluation of the Eco-Pick lift assist: A device designed to facilitate product selection tasks in distribution centers

Developing lift assist devices to aid workers performing case tasks in distribution centers has been challenging given the movement of workers through a distribution facility. The objective of this work was to biomechanically evaluate a lift assist that can be integrated with pallet jacks and thereby move through the facility with the workers. Twelve participants transferred 16.4 kg boxes from one pallet to another manually and using the Eco-Pick lift assist. Electromyographic (EMG) activities were measured bilaterally in the Bicep, Deltoid, Latissimus Dorsi, and Erector Spinae muscles. The analysis showed that the 90th percentile normalized EMG values were significantly reduced in 4 of the 8 muscles sampled when using the Eco-Pick. Likewise, the 50th percentile normalized EMG data were significantly reduced for the Erector Spinae, Latissimus Dorsi, and the Bicep muscles when using the assist. For some of the muscles the advantage of the Eco-Pick was dependent upon the initial lift height or the placement height. Overall, it appears that this evaluation of the Eco-Pick's efficacy indicates that the device has the potential to be an effective means of reducing worker exposure to risk factors associated with back and shoulder injuries experienced during manual palletizing tasks found in distribution centers.

Development of a robot simulator using a modified trajectory generation algorithm oriented towards the palletizing task

Palletizing tasks are necessary to promote the efficient storage and shipping of boxed products. These tasks, however, involve some of the most monotonous and physically demanding tasks in the factory. Thus, many types of robot palletizing systems have been developed, although many robot motion commands still depend on the teach pendant. That is, the operator inputs the motion command lines one by one. This is very troublesome and, most importantly, the user must know how to type the code. To solve these problems, this paper proposes a combination of an optimized pallet pattern generation algorithm, an industrial robot simulator, and a modified trajectory optimization algorithm. To integrate these modules and to define the position of the boxes, the robot and its peripherals, as well as the system layout and its coordinates, were defined. Finally, the proposed path generation algorithm, the ‘overlap method’, was tested in the designed offline programming S/W, and its computational efficiency was proven.

로봇 팔레타이징 시뮬레이터를 위한 적재 패턴 생성 및 시변 장애물 회피 알고리즘의 제안

Palletizing tasks are necessary to promote efficient storage and shipping of boxed products. These tasks, however, involve some of the most monotonous and physically demanding labor in the factory. Thus, many types of robot palletizing systems have been developed, although many robot motion commands still depend on the teach pendant. That is, the operator inputs the motion command lines one by one. This is very troublesome and, most importantly, the user must know how to type the code. We propose a new GUI(Graphic User Interface) for the palletizing system that is more convenient. To do this, we used the PLP "Fast Algorithm" and 3-D auto-patterning visualization. The 3-D patterning process includes the following steps. First, an operator can identify the results of the task and edit them. Second, the operator passes the position values of objects to a robot simulator. Using those positions, a palletizing operation can be simulated. We chose a widely used industrial model and analyzed the kinematics and dynamics to create a robot simulator. In this paper we propose a 3-D patterning algorithm, 3-D robot-palletizing simulator, and modified trajectory generation algorithm, an "overlapped method" to reduce the computing load.

The effect of pallet distance on torso kinematics and low back disorder risk

Intervention research for prevention of occupational low back injuries has focused on the effects of reducing extreme torso flexion and the external moment. Little is known about prevention strategies for torso twisting and lateral bending. The objective of this study was to assess the effect of pallet distance with regard to a constant lift origin on the torso kinematics and a measure of low back disorder risk. Fifteen male participants transferred 11.3 kg boxes from a constant origin to six different regions on a pallet. Two pallet distances with regard to the lift origin were investigated. ANOVA indicated that increasing the pallet distance resulted in increases in torso kinematics (velocities and accelerations) as well as a measure of risk of low back disorder. The increases in torso kinematics (e.g. twisting and lateral awkward postures and bending velocities) occurred mostly at the lower height regions on the pallet. It is concluded that increasing the pallet distance with regard to the lifting origin, with the intention to influence the participant to take a step during a palletizing task does not appear to be an effective intervention strategy to reduce the risk of low back disorder associated with torso kinematics.

Comparison of the sensitivity of three psychophysical techniques to three manual materials handling task variables

The primary objective of the study was to determine the absolute and relative sensitivity of three psychophysical techniques of physical stress determination (Borg's Scale, the Visual Analogue Scale, and the Body Part Discomfort Rating) in varied manual materials handling tasks. Ten young male and ten young female university students lifted a 22 kg box onto a shelf 132 cm high. The industrial-size box (40 x 20 x 20 cm) had two cut-out handles on the top middle of the two side panels. The shelf had restricted access allowing 5 and 10 mm clearance for access. The palletizing was performed in sagittal and 45° asymmetrical postures under unlimited, 90% and 80% of stature headroom. The subjects lifted the box six times per minute for 5 min. After every lift the load was automatically delivered to the starting position. The 12 experimental conditions were randomized. Following each 5-min palletizing task the subjects were required to assess the immediately foregoing task on Borg's scale, the Visual Analogue scale, and a Body Part Discomfort Rating scale. All data were subjected to statistical analysis including multivariate analysis of variance and Scheffe post-hoc multiple range test. Borg's scale was able to discriminate between two clearances, two symmetries and the three headrooms (p< 0.001). The Visual Analogue scale was able to differentiate between two symmetries and three headrooms (p< 0.01), but the Body Part Discomfort Rating could not differentiate between any task conditions. Based on the study it is suggested that an appropriate match between the task demands and the basis of a psychophysical tool is essential for valid and reliable information.

Reduction of spinal loading through the use of handles

A study was performed to investigate how different types of handle coupling affect the loading on the spine. Ten male grocery item selectors performed a laboratory simulation of a warehouse palletizing task. Participants transferred the cases from a pallet in a storage bin to the destination pallet. The trunk motions and muscle activities were monitored by a Lumbar Motion Monitor (LMM) and electromyographic electrodes, respectively, and used as input to an EMG-assisted biomechanical model. The results of the study revealed that the presence of handles reduced the complex loads on the spine. This was particuarly true when lifting to the lowest positions of the pallet, where the highest forces occurred. It was determined that the maximum spinal compression forces were reduced by an average of 6.8% when handles were added to the cases. The presence of handles affected the moments imposed on the trunk in the lower regions on the pallet, indicating a difference in lifting style and/or more sagittal flexion. The results of this study suggest that the multiplier for handle coupling in the 1991 NIOSH Revised Lifting Equation was appropriate for higher lifts (at 133.8 cm), but needs to be more protective for ‘poor’ coupling conditions with lower vertical heights, which are the most common in industry. Based on these results, it is recommended that handles be designed into the cases that are commonly lifted from low levels in warehousing and other manual materials handling situations.

A Case Study of Factors Influencing the Effectiveness of Scissor Lifts for Box Palletizing

This case study investigated why scissor lifts installed by a meat processing plant for a conveyor-to-pallet manual materials handling (MMH) task were not used by employees, despite apparent workplace improvement. A lumbar motion monitor (LMM) was used to assess tri-axial trunk motions and the velocities and accelerations associated with the job. Three palletizing conditions were evaluated: at the floor, at a scissor lift, and at a modified scissor lift. Front and back box positions at three levels of the floor palletizing task and three box positions at the scissor lift were assessed. Analysis combined observations, probability of “high risk” for low back disorder (LBD), and the National Institute for Occupational Safety and Health 1991 Work Practices Guide lifting index. Results showed the original scissor lift workplace had a higher probability of LBD risk than floor palletizing due to subtle barriers. These included a raised edge on the conveyor and a railing around the scissor lift where boxes were placed, which increased maximum sagittal flexion and maximum lateral velocity during lifting. However, the NIOSH lifting index suggested an improvement of the scissor lift over the floor palletizing task, but the guideline is not sensitive to dynamic factors. After modifications to the barriers, MMH at the scissor lift workstation had a risk comparable to the medium level of floor palletizing. The changes reduced the higher risk associated with the identified installation problems and the low level of MMH at the floor. Results showed that measurement of dynamic factors provided greater sensitivity to the subtleties of workstation barriers and illustrated the effect of small barriers on task performance and associated risk. Further studies of trunk motions with alternative mechanical aids and different starting heights in MMH are indicated.

A Case Study of Factors Influencing the Effectiveness of Scissor Lifts for Box Palletizing

This case study investigated why scissor lifts installed by a meat processing plant for a conveyor-to-pallet manual materials handling (MMH) task were not used by employees, despite apparent workplace improvement. A lumbar motion monitor (LMM) was used to assess tri-axial trunk motions and the velocities and accelerations associated with the job. Three palletizing conditions were evaluated: at the floor, at a scissor lift, and at a modified scissor lift. Front and back box positions at three levels of the floor palletizing task and three box positions at the scissor lift were assessed. Analysis combined observations, probability of “high risk” for low back disorder (LBD), and the National Institute for Occupational Safety and Health 1991 Work Practices Guide lifting index. Results showed the original scissor lift workplace had a higher probability of LBD risk than floor palletizing due to subtle barriers. These included a raised edge on the conveyor and a railing around the scissor lift where boxes were placed, which increased maximum sagittal flexion and maximum lateral velocity during lifting. However, the NIOSH lifting index suggested an improvement of the scissor lift over the floor palletizing task, but the guideline is not sensitive to dynamic factors. After modifications to the barriers, MMH at the scissor lift workstation had a risk comparable to the medium level of floor palletizing. The changes reduced the higher risk associated with the identified installation problems and the low level of MMH at the floor. Results showed that measurement of dynamic factors provided greater sensitivity to the subtleties of workstation barriers and illustrated the effect of small barriers on task performance and associated risk. Further studies of trunk motions with alternative mechanical aids and different starting heights in MMH are indicated.

Adaptive Reinforcement Learning and Its Application to Robot Compliance Learning

A new learning algorithm for connectionist networks that solves a class of optimal control problems is presented. The algorithm, called Adaptive Reinforcement Learning Algorithm, employs a second network to model immediate reinforcement provided from the task environment and adaptively identities it through repeated experience. Output perturbation and correlation techniques are used to translate mere critic signals into useful learning signals for the connectionist controller. Compared with the direct approaches of reinforcement learning, this algorithm shows faster and guaranteed improvement in the control performance. Robustness against inaccuracy of the model is also discussed. It is demonstrated by simulation that the adaptive reinforcement learning method is efficient and useful in learning a compliance control law in a class of robotic assembly tasks. A simple box palletizing task is used as an example, where a robot is required to move a rectangular part to the corner of a box. In the simulation, the robot is initially provided with only predetermined velocity command to follow the nominal trajectory. At each attempt, the box is randomly located and the part is randomly oriented within the grasp of the end-effector. Therefore, compliant motion control is necessary to guide the part to the corner of the box while avoiding excessive reaction forces caused by the collision with a wall. After repeating the failure in performing the task, the robot can successfully learn force feedback gains to modify its nominal motion. Our results show that the new learning method can be used to learn a compliance control law effectively.

線形計画を用いた組立作業に対するコンプライアンス設定法

A new approach is presented for designing compliance matrices for assembly tasks using linear programming. Compliance plays a significant role in performing manipulation tasks which involve contact with rigid objects such as assemblies. However, there is no useful method for designing a compliance matrix which' can be adapted to each task.In this paper, compliance is treated as a linear mapping from a force to a corrected motion and is constructed on the basis of geometric information. First, a basic framework of assembly task strategies is described. This is a hierarchical framework based on primitive motions. The primitive motion is defined as a motion that accomplishes a transition from an initial state to a goal state, given a single compliance matrix and a single motion command. The presented method designs compliance matrices for primitive motion. Next, the three conditions required of an appropriate compliance matrix, stability, feasibility, and error correctability, are discussed. These conditions are then formulated as a set of linear inequalities with respect to compliance matrix elements, and the compliance matrix design problem is reduced to a linear programming problem, with this set of inequalities as constraints. Finally, the usefulness of this method is shown by a box palletizing task.