Abstract
BACKGROUND:
Depletion of groundwater has increased the depth of bore wells drilled to meet water requirements. Water from these bore wells are accessed by operating submersible pumps with multi-stages. The number of stages of submersible pump directly depends upon the depth of bore well. Health-related risk such as Musculoskeletal Disorders (MSDs) arises during the manual assembly of these pumps in an awkward posture.
OBJECTIVE:
The objective of this paper is to eliminate or reduce risk in the manual assembly of submersible pumps using virtual postural analysis. Secondary objective is to redesign the existing workstation incorporating ergonomic interventions.
METHODS:
It reports a case study in which postural analysis is carried out with the help of Rapid Upper Limb Assessment (RULA) method.
RESULTS:
It is observed that inserting the stage casing weighing around 1.5 Kg through top end of the shaft is the awkward working posture involving a flexion of the arm about 152°. RULA score (7) indicate a high level of risk exists in the workstation, calling for investigate and change the working posture immediately. In order to reduce this risk, a redesigned workstation has been proposed.
CONCLUSION:
This study results help in minimizing the awkward postures, thereby reducing the health-related risk. This will also increase the efficiency of the worker and his productivity.
Keywords
Introduction
Water is a fundamental necessity among natural resources. The foremost sources of water embody rivers, rain and groundwater. The streams meet the water necessities just in districts around waterway banks and places where water circulation channels are accessible. As of late, an unnatural weather change, populace development, urbanization, and conjecture disturbance to rain pattern have made individuals depend more on groundwater for their agribusiness and residential needs. A technical report from the Central Ground Water Board [1], Republic of India (2008) revealed the subsequent information: the restricted accessibility and increased demand has resulted in depletion of existing spring water level by 5 meters/year. This brought about penetrating bore wells to depths of 100–250 meters (m) and at times, even 300 m taking into account accessibility of the water table. Submersible pumps with multi-stages are utilized in these bore wells. Each stage of a submersible pump has the ability to draw water from a depth of around 10 m [2].
A small-scale submersible pump manufacturing industry lacks automation facilities, because of investment constraints and therefore assembled manually [3]. This manual assembly is finished with jigs, tools like hammers, and spanners over tables known as workstations. These workstations ought to incorporate ergonomic principles to avoid awkward operating postures which can cause health related risks like Musculoskeletal Disorders (MSDs) [4–7]. MSDs are injuries and disorders of soft tissues like tendons, ligaments, fibrous tissues, synovial membranes, muscles, nerves and blood vessels [8].
In order to identify and eliminate or reduce the health-related risks, working postures involved in the task has to be analyzed. Rapid Upper Limb Assessment (RULA) is a posture analysis tool developed by McAtamney and Corlett [9] to analyze the working postures. This tool needs no special instrumentation in providing a quick assessment for postures of the neck, trunk, and upper limbs in conjunction with muscle function and external loads experienced by the worker. A coding system is used to produce action lists which indicate the level of intervention requisite to minimize the risk. The popularity of this tool has made the Computer Aided Design (CAD) packages to incorporate it as a module [10–12].
Literature related to working postures, virtual ergonomic assessment with CAD, and Digital Human Modelling (DHM) are discussed in this section. English et al. [13] reported that higher rates of wrist flexion/extension and prolonged period of shoulder rotation with elevated arm increase the risk of upper limb soft tissue disorders. Statically holding components with palm and fingers supplement the prevalence of disorders [14]. In submersible pump assembly, these postural issues can occur when placing the components in proper position. During stage assembly of the pump, hands and arms of the worker are above shoulder or acromion height and it was categorised as overhead work [15]. Overhead work consists of awkward postures which may generate discomfort and pain to shoulders and leads shoulder disorders [16]. These postural issues can be addressed using proper ergonomic interventions in order to reduce the likelihood of MSDs [4–7]. Literature indicates that DHM was extensively used to develop such ergonomic interventions in workstations [10, 17–20], retail spaces [11], and products [21, 22]. Doi et al. [23] used DHM and virtual workstations to analyse postural behaviour of sewing machine operators. DHM is also reliable in detecting awkward postures and virtual assessments in workstations [24].
The main advantage of this virtual assessment technique is that the working posture can be analyzed in a CAD environment with no real-time implementations [10–12, 17–24]. The concepts from the above-published papers inspired to carry out the virtual postural assessment of an awkward working posture concerned with manual assembly of submersible pumps in small-scale manufacturing industries. The main objective of this paper is to eliminate or reduce MSD risk in the manual assembly of submersible pumps by assessing the level of risk in existing workstation using virtual postural analysis and proposing a newly redesigned workstation incorporating ergonomic intervention. So as to satisfy the objective, a case study in a small scale manufacturing industry was conducted.
Method
As a pre-requisite of the DHM based workstation evaluation, a preliminary study was conducted to identify musculoskeletal problems with a self-report. Participants were thirty male assembly workers from various submersible pump manufacturing industries in south India. The stature of these workers provided a median value of 165 cm with an inter-quartile range of 7.75 cm. Cornell Musculoskeletal Discomfort Questionnaire was used in the study for the assessment [25]. In this self-report, 100% of workers were reported shoulder discomfort with 88.8% frequency of occurrence, 60% severity and 62% interference. 83.3% of workers were reported wrist discomfort and 70% of workers were reported discomfort in neck while discomfort for the lower parts of the body is lesser than this statistics. Also prevalence percentage based on MSD impact score was found higher in shoulder and wrist. These results indicate the requirement of conducting a detailed study in existing workstation to eliminate or reduce MSD risk. In order to obtain the desired objective, a methodology is adapted as shown in the flow chart (Fig. 1).

Flow chart.
The first step is to collect data in a small-scale submersible pump manufacturing industry. This involves collecting data from an existing workstation in the form of photographs and sketches. An awkward posture involved in the assembling task is also identified from the various working postures. The second step is to create a virtual workstation with the integration of CAD and DHM using appropriate anthropometry data.
The third step is a virtual postural assessment using RULA tool. This assessment examines the risk factors such as a number of movements, static muscle work, force and working posture. All these factors combine to provide a final score that ranges from 1 to 7. The final score is displayed by a colored zone. The zone color changes from green to red according to the final score. The various final score levels, color coding, and their indications are shown in Table 1. Based on the posture assessment results, it is decided whether to accept the existing posture or change the working posture by redesigning the workstation incorporating ergonomic considerations.
RULA action levels
As indicated in the methodology, a data collection regarding components involved in assembly, anthropometry, workstation, posture were made in the assembly workstation of a small-scale submersible pump manufacturing industry. The submersible pump incorporates outer components like the suction end, stage casings and non-return valve (NRV) as shown in Fig. 2. It also includes inner components such as impellers, shaft, coupling, sleeve and bushes. The outer set of components is to be assembled leakage proof at every stage, with ‘interference fit’. This is made certain by hammering of components one over another. The inner set of components is assembled with ‘clearance fit’. The proper positioning of inner components is not easy in horizontal orientation. Due to this constrain, the pump assembly is done only in vertical orientation. This ensures proper functioning of the assembled product.

submersible tube well pump with twenty stages.
The dimensions of the workstation, pump components and tools were measured and recorded as sketches. The anthropometries of the workers involved in this assembly task were measured using Lafayette anthropometer and noted. The various working postures involved in this assembly task are also recorded using photographs. During this study, it is observed that a worker has to fix the shaft of the pump in a jig attached to a table and assemble the components through the top end of the shaft. Here the entire assembly work is done with reference to the center shaft. So, each time to assemble a component, the worker has to access the component from the table and assemble it through the top end of the shaft. This mainly involves working postures with hands above shoulder level as shown in Fig. 3. Among the various working postures, assembly of stage casings weighing around 1.5 kg is the awkward posture involved in this task. The commonly manufactured submersible pumps are about 12 stages, 20 stages, 25 stages and 30 stages. The annual demand for 20 and 25 staged pumps together comes around 80% while the demand of 30 staged pumps comes only at 2%. When the number of stages increases, the height of the shaft also increases and the number of components to be inserted also increases. The height of the center shaft becomes a problem during this assembly task.

Assembly of twenty stage pump.
The CAD models of the components in the existing workstation were made based on the recorded sketches and then imported into ‘Ergonomics design & analysis’ module of CATIA-V5 software. These components are arranged to obtain the virtual environment of the existing workstation. This module includes four sub-modules as ‘Human builder’ to create DHM called manikins, ‘Human measurements editor’ to edit the anthropometry of manikins, ‘Human posture analysis’ to analyze manikin postures, and ‘Human activity analysis’ to analyze how a manikin interacts with objects in its virtual environment.
Five default populations: American, Canadian, French, Japanese and Korean are available in its database to create a manikin. But, while designing the workstation for Indian workers, the manikin used must correspond to the Indian population. The anthropometric data of Indian population developed by Chakrabarti [26] is used to meet this requirement. Also, only male workers are engaged in pump assembly work and hence, the anthropometric data corresponding to male population is considered throughout the study. From the preliminary study, it was found that the median value of stature in assembly workers is nearer to the 50th percentile value of Indian population (i.e., 164.8 cm). The measured anthropometric dimensions such as stature, cervical height, acromion height, chest height, waist height, crotch height, dactylion height, chest breadth and waist breadth of existing three assembly workers in the present industry were verified. It was found that the values are nearer to the data of 50th percentile Indian population [26].
Moreover, in submersible pump assembly work, the top end of the shaft and work table height determines the assembly height. The assembly height is fixed for each type of pump. If the DHM adapts 5th percentile of population in order to accommodate maximum assembly height for smaller workers, it makes some bending posture to attain minimum assembly height for the 95th percentile population. Similarly, if DHM adapts 95th percentile population in order to accommodate minimum assembly height, it makes unable to reach the maximum assembly height for 5th percentile population. It reveals that in order to suit this assembly work for a desired population, a safer and adaptable range of working height should be determined based on the fixed assembly height. To determine a basic solution to this problem, DHM adapts anthropometric data of 50th percentile population which matches the data of assembly workers [12]. Assigning these workers to pump assembly job confirms concept of providing suitable worker to work [27, 28]. The manikin is generated using ‘Human builder’ module of the software and edited in ‘Human measurements editor’ using the anthropometric data of 50th percentile Indian population [26]. This manikin is incorporated with the virtual environment of the existing workstation for 20 staged pump assembly. Figure 4 (a) shows the virtual workstation with imported CAD models like the table, jig, the suction end, shaft, stage casings, impeller, sleeve and tools like hammers incorporated with the manikin.
In this assembly work, vertical height is the most important factor which affects the working posture. Some of the vertical heights that will have an impact on the working posture of this manual assembly are table height, shaft height, and height of the manikin. These heights are shown in Fig. 4 (b).

Virtual existing workstation; (a) Components, (b) Dimensions.
In the existing workstation setup, the case study revealed that inserting the stage casing weighing around 1.5 kg through the top end of the shaft is the awkward working posture involving a flexion of the arm about 152° as shown in Fig. 5. This working posture is repeated less than 4 times per minute, which means the posture is intermittent between static and repeated postures. Hence, the virtual posture assessment is performed separately for left and right sides for this working posture with the posture as intermittent and load value as 1.53 kg (Weight of the stage casing).

Awkward working posture in existing workstation.
Figure 6 shows the RULA results of left and right sides with the segmental scores and the final score. The segments of the manikin are color coded representing the segment scores.
In this analysis, the upper arm gets a score of ‘6’, forearm gets a score of ‘2’, a score of ‘2’ for the wrist, ‘1’ for wrist twist, ‘4’ for the neck, ‘1’ for trunk, and ‘1’ for legs. Here, the muscle and force/load scores are ‘0’. The score A is ‘8’ and score B is ‘5’. The final score is ‘7’ (i.e., investigate and change immediately) for both right side and left side which is a critical value. Hence, this awkward working posture must be changed immediately by redesigning the workstation incorporating ergonomic considerations to reduce the health-related risk involved in this task.

RULA for existing workstation.
The proposal focuses on redesigning the workstation to fit the worker and eliminate awkward postures. The objective is to improve the overall safety and performance of employees through the application of sound workstation design and eliminate or reduce health-related risks. The following section describes the redesign of the workstation.
When the number of stages of pump increases, vertical height of the top end of the shaft increases. All these pumps with a different number of stages are assembled in the same table. It in turn increases the assembly reach height of workers and develops the awkward working postures due to overhead work. Hence, this redesign mainly focuses on reducing the overhead work. In redesign also 50th percentile population parameters and 20 staged pump are adapted for obtaining a comparison with existing method.
The redesign is done in three iterations as shown in Fig. 7. In order to reduce overhead work in the assembly, in the first iteration of redesign, the total assembly height is divided in to three parts. The worker is provided with a three stepped platform to do the assembly. A rack arrangement is also provided in the work table to stock the components and tools. The RULA analysis provides safer score at higher stages of assembly. In second iteration, table is modified in to a semicircular racked one to provide easiness in the handling of components. Finally in third iteration, three stepped platform is replaced with single platform and a pneumatic actuation system is introduced to assist the assembly. The racked table is changed in to semi-circular table.

Iterations in redesign of workstation.
The modified workstation is based on the third iteration. The dimensional details of modified workstation are shown in Fig. 8. The pneumatic actuation system in this design is with a hollow fixture of 10 mm thickness and 900 mm extension height from the ground level. It works as a base for the assembly. It consists of two pneumatic cylinders on both sides of the fixture to withstand the weight of the assembly. In order to obtain minimum reach of 610 mm for the worker, a platform level of 300 mm is introduced. A jig is fixed above the ground level at a height of 450 mm for holding the shaft. 450 mm is the height of pneumatic system after maximum retraction (i.e., taken as half of the extension height). The shaft height in the pump is 1265 mm from the base of the jig. The maximum reach height of the worker is 1415 mm from the platform level. It is above the acromion height of 50th percentile population and reaches the cervical height [26].

Modified workstation.
In order to validate the worker’s compatibility in the workstation to access the components, the work envelope of the manikin is generated and analyzed. Figure 9 shows the work envelope of the manikin in the redesigned workstation. The manikin has a vertical envelope of 1531 mm, 1670 mm from right to left, 1460 mm from front to back and a volume of 3.005 m3. It also revealed that the maximum surface of the workstation is covered by the work envelope of the DHM.

Work envelope; (a) Top view, (b) Front view.
During assembly, the pneumatic cylinders will be initially in extension condition. The modified workstation provides a working range of 805 mm between minimum and maximum reach heights of the worker. The assembly of 20 staged pump can be completed with two phases of operation. During the first phase, worker has to fix the suction end with a height of 95 mm in the fixture and insert the shaft into the jig by passing it through the hole in the fixture. Then, the assembly of stages with stage casings is done as shown in Fig. 10 (a). Each stage has a height of 55 mm. Therefore, a worker can do 12 stages (660 mm) of assembly within this working range. The second phase starts with the retraction of the pneumatic cylinders which moves down the completed assembly through the shaft as shown in Fig. 10 (b). This enables to do the assembly of remaining eight stages (440 mm) within the working range.

States in assembly; (a) assembly of the stage casing, (b) Pneumatic cylinders retraction.
The RULA is performed for the working postures of the manikin in the proposed workstation. The working postures involve fixing the suction end in fixture and inserting the stage casing through the top end of the shaft. By standing on the platform, fixing the suction end in a fixture is done when the pneumatic cylinders are in extension condition. This working posture involves a flexion of 5° for the upper arm and 34° for the forearm. The RULA is performed for this working posture as shown in Fig. 11.

RULA for fixing suction end.
The results were obtained with the upper arm score of ‘1’, forearm score of ‘2’, a score of ‘1’ for wrist, ‘1’ for wrist twist, ‘1’ for a neck, ‘2’ for trunk, and ‘1’ for legs. Here, the muscle and force/load scores are ‘0’. The score A is ‘2’ and score B is ‘2’. The final score is obtained as ‘2’ (i.e., acceptable). This implies that this working posture is a safer one and can be adopted by the worker.
After fixing the suction end in the fixture, the shaft is inserted into the jig. Now, the person has to access the components placed on the table and assemble it by inserting them through the top end of the shaft. The assembly is done up to a certain number of stage casings and then the pneumatic cylinders are made to retract. This is done to lower the fixture and then assemble remaining components. This working posture involves a flexion of 19° of the upper arm and 128° of the lower arm. The RULA is performed for this working posture as shown in Fig. 12.

RULA analysis for assembling the stage casing.
The results were obtained with the upper arm score of ‘2’, forearm score of ‘2’, a score of ‘1’ for the wrist, ‘1’ for wrist twist, ‘1’ for the neck, ‘2’ for trunk, and ‘1’ for legs. Here, the muscle and force/load scores are ‘0’. The score A is ‘3’ and score B is ‘2’. The final score is obtained as ‘3’ (i.e., acceptable). This implies that this working posture is also safe and can be adopted by the worker.
The segmental scores and final score for the postural assessment of existing and proposed workstations are compared in Fig. 13. The upper arm score is reduced from a maximum value of ‘6’ to the least value of ‘2’, the wrist score gets reduced from ‘2’ to ‘1’, the neck score from ‘4’ to ‘1’, score A is reduced from a critical value of ‘8’ to ‘3’, and score B from a value of ‘5’ to a new value of ‘2’. This comparison reveals how the risk level in each segmental part has been reduced by eliminating awkward postures. The final score obtained as ‘7’ for the existing case is now reduced to a new value of ‘3’.

Comparison of segment scores and final scores.
Mohan et al. [10] applied RULA tool to evaluate the working posture of foundry workers and reduced the final score value from ‘7’ to an acceptable value of ‘3’ by workstation redesign. Kushwaha and Kane [12] also attained similar result in the redesign of crane cabin for seated workers. In an ergonomic intervention study at retail stall RULA tool was used to analyze postures and reduced the final score from ‘6’ to a new value of ‘3’ [11]. The final score obtained in the present study is also in congruent with these results.
The proposed design is flexible to suit for any desired population. The design works on the principle of attaining maximum reach of the worker in the assembly by setting base height of the jig and minimum reach of the worker by fixing worker’s platform height. This principle can be adapted to the conventional design, in which the maximum reach of 5th percentile population and minimum reach of 95th percentile population determines the working range. The design can provide a working range of 618 mm based on the maximum reach of 5th percentile population up to cervical level, in order to reduce overhead work [26]. A safe design can be done based on this working height and the working range suits for 90% Indian population. It is evident that, this working range is lesser than that of the modified design explained in this paper (i.e., 805 mm). It is also evident that, accommodating more population in design causes decrease in the working range. Smaller working ranges will increase the number of retractions of pneumatic cylinder and thereby increase in the cycle time. Also the design can be suited for different staged pumps by adapting the same principle. The cycle time increases with number of stages of pump while the risk remains same. However, for each staged pump it requires to have separate workstations.
The anthropometric data other than stature were verified with all the pump assembly workers in the studied industry, but there were only three workers existed in the industry. Also, the modified design involves a component of overhead work, as the maximum reach height in the design is above acromion height. The design is developed as a basic feasible solution for the risks due to awkward postures. Further research is needed to optimize the design. It is difficult to eliminate overhead work and its risks due to awkward postures from industrial tasks, but it can be minimized [29].
In this paper, a case study of an assembly work in a small-scale submersible pump manufacturing industry was presented. An awkward working posture involved in this assembly work was also identified. A virtual workstation was created with the help of CAD model and DHM. The 50th percentile Indian anthropometry data was used for this purpose. In this virtual workstation, a postural assessment is made for the identified awkward working posture using RULA tool. This virtual posture assessment revealed a final score of ‘7’ which clearly indicated that an investigation to change the working posture is needed immediately. The workstation was redesigned incorporating ergonomic considerations to eliminate awkward posture which reduced the final score value to ‘3’. Hence, this working posture is acceptable and the proposed workstation can be implemented in the small scale submersible pump manufacturing industries to avoid awkward postures. During this study similar problems are also observed in valve assembly and motor assembly workstations. The basic problem behind these assemblies remains the same (i.e. vertical height of the assembly). The methodology proposed in this work can be adopted in the above said workstations to reduce the health-related risks. At present pneumatic fastening system is used in the final assembly of submersible pumps and workers are comfortable with the same. Hence it seems that modified workstation can easily be adapted for this task with the approval from management. Future direction of research could include design optimization by providing adjustability in height of the jig base and worker’s platform. Such design would improve the accessibility of workers to the assembly and it would further reduce the risk of MSDs.
Conflict of interest
None to report.
Footnotes
Acknowledgments
Authors would like to acknowledge the efforts of reviewers and editors in bringing out this study to public domain. Authors are also grateful to acknowledge the support of the management, staff and workers of PSG Industrial Institute, Coimbatore, India.
