MODELLING, SIMULATION AND ERGONOMIC STANDARDS AS SUPPORT TOOLS FOR A WORKSTATION DESIGN IN MANUFACTURING SYSTEM

Antonio Cimino(a), Giovanni Mirabelli(b)

(a) (b) Modeling & Simulation Center - Laboratory of Enterprise Solutions (MSC-LES) Mechanical Department, University of Calabria, 87036 Rende (CS), Italy

(a) acimino@unical.it, (b)g.mirabelli@unical.it

ABSTRACT It is the intent of the work to achieve the ergonomic effective design for a workstation belonging to a manufacturing system operating in the field of mechanical parts production. To this end, the NIOSH 81, NIOSH 91, Burandt Schultetus and OWAS analysis are accomplished for developing an improved workstation configuration in terms of interaction between humans and their working environment. Moreover the authors adopt an advanced approach based on Modeling & Simulation (M&S) in order to implement a three dimensional environment for recreating with satisfactory accuracy the real workstation and for detecting ergonomic problems that otherwise would be difficult to detect. In conclusion a final workstation configuration is proposed and significant ergonomics improvements are achieved.

Keywords: workstation effective design, ergonomic standards, modeling, simulation.

1. INTRODUCTION An overview of the state of the art, starting from the second half of the 1990s, reveals that manufacturing systems continuously provide challenging problems for researchers and scientists working in the applied ergonomics field. A number of different research works and scientific approaches have been proposed, trying to achieve an ergonomic effective design of the workstations belonging to a manufacturing system.

In the late ’90, the ergonomic effective design was mostly supported by videotaping systems used for data collection, i.e. videotape of the work method being used in the workplace (Das and Sengupta 1996; Scott and Lambe 1996; Engstrom and Medbo 1997; Vedder and Hellweg 1998; Kadefors and Forsman 2000; Neumann et al. 2001).

In addition, several authors propose an approach based on a single ergonomic performance measure (i.e. lift index, energy expenditure measure, work postures etc.) due to the application of a specific ergonomic standard such as the Ovako Working Posture analysis System (OWAS), the Burandt-Schultetus analysis, the NIOSH 81 and NIOSH 91 equations (NIOSH stand for

National Institute for Occupational Safety and Health), the RULA method (RULA stands for Rapid Upper Limb Assessment), the Garg analysis. Further information about the cited ergonomic standards can be found in the Niosh Technical Report 81-122 (1981), the Scientific Support Documentation for the Revised 1991 NIOSH Lifting Equation (1991), Waters et al. (1994), Kharu et al. (1981), Schultetus (1980) and Garg (1976), McAtamney and Corlett (1993). Examples of research works based on a single ergonomic performance measure are reported in Carrasco et al. (1995), Grant et al. (1997) Van Wendel de Joode et al. (1997), Mital and Ramakrishnan (1999), Temple and Adams (2000), Lin and Chan (2007), Waters et al. (2007).

In order to achieve relevant ergonomic improvements some authors propose an effective ergonomic design based on the integration of different ergonomic standards. Wright and Haslam (1999) investigate manual handling risks within a soft drinks distribution centre using the OWAS postural analysis and the NIOSH equations. Russell et al. (2007) compare the results of different ergonomic standards for evaluating ergonomic risks in lifting operations. Jones et al. (2005) present an examination of three common pub occupations (bartending, waitressing and cooking by means of RULA method and NIOSH Lifting Equation.

Another critical issue, other than the choice of the ergonomic performance measure (i.e. single or multiple) or the methods used for data collection (i.e. video type system), is whether the ergonomic standards have to be applied directly in the real system. Usually the application of the ergonomic standards in the real system requires huge amount of money and time for testing all the workstations configurations, work assignments, work methods, etc. Consequently, researchers and practitioners very often use Modelling & Simulation (M&S) as decision support tool for choosing correctly, for understanding why, for diagnose problems and explore possibilities (Banks, 1998).

Over the years the M&S approach has became more and more appealing thanks to the numerous advantages such as the possibility to study ergonomic issues at the earliest stages of design in order to avoid

788

potential future ergonomic redesign in the real-world system. Feyen et al. (2000) propose a PC-based software program (based on the integration of a Three-Dimensional Static Strength Prediction Program, 3DSSPP, for biomechanical analysis with a widely used computer-aided design software package, AutoCAD). As consequence, the authors are able to study ergonomic issues during the design phase taking into consideration different design alternatives.

Chang and Wang (2007) propose a method for conducting workplace ergonomic evaluations and re-design in a digital environment with the aim of preventing work-related musculoskeletal disorders during assembly tasks in the automotive sector.

Longo et al. (2006) use M&S in combination with ergonomic standards and work measurement for the effective design of an assembly line still not in existence. The authors propose a multi-measures approach with the aim of obtaining a different work assignment to each workstation, better line-balancing and better ergonomic solutions.

Santo et al. (2007) propose an ergonomic study on working positions in a manufacturing company (by using the simulation software eM-Workplace) and providing, as result, remarkable ergonomic improvements.

The main goal of the paper is to achieve the ergonomic effective design of a real workstation belonging to a manufacturing system operating in the field of mechanical parts production. To this end, several ergonomic standards are accomplished for developing an improved workstation configuration in terms of interaction between humans and their working environment. Moreover the authors adopt an advanced approach based on Modelling & Simulation (M&S) in order to implement a three dimensional environment for recreating with satisfactory accuracy the real workstation and for detecting ergonomic problems that otherwise would be difficult to detect.

Before getting into details of the study let us give a brief overview of each section of the paper. Section 2 describes the manufacturing process. Section 3 gives specific details on the simulation model implementation and validation. Section 4 presents the ergonomics standards accomplished for the effective ergonomic design. Section 5 presents the simulation results. Section 6 shows the effective ergonomic redesign of the workstation under consideration (final configuration). The last section reports the conclusions that summarize the scientific contribution of the work and the research activities still on going.

2. THE MANUFACTURING PROCESS The research work focalizes on the ergonomic effective design of a workstation belonging to a manufacturing system specialized in the field of mechanical parts production. The manufacturing plant is located in the South part of Italy and covers a surface of 11000 square meters. It manufactures a mechanical component for

agricultural machineries engines on behalf of third parties and consists of three different areas:

• The raw materials storehouse: a basic mechanical component is stored in pallets loaded into pallet racks. The pallets and the mechanical components are moved by means of a system of automated conveyors, automated storage and retrieval machines. The storage area is 11 meters high and covers a surface of 2000 square meters;

• The production area: it consists of three workstations: the flange workstation, the screws workstation and the assembly workstation. The flange workstation products the flanges being assembled to the mechanical component stored in the raw materials storehouse. It employs 4 operators and contains 4 motorized ball control lathes characterized by the same productive process. The workstation layout covers a surface of 2500 square meters. The Screws workstation manufactures 10 different screws types; in particular the hexagonal screws 4 cm high are used for fixing together the mechanical component and the flange, while the other screws types represent, for the manufacturing system, final products directly sold to the customers. The workstation employs 5 operators and it is made by a broaching machine, a slotting machine, a machine for worm screws, a numerically controlled machine for worm screws with robot and a shaving machine. The workstation layout covers a surface of 2500 square meters. The Assembly workstation assembles the mechanical component, the flange and the screws in order to obtain the main final product for the enterprise. The assembly activities are performed during all the day by three experts operators (each operator works eight hours per day). The plant layout covers a surface of 2000 square meters.

• The final products warehouse: the assembled workpiece and the screws manufactured in the production department are stored in pallet loaded in pallet racks. As the Raw materials storehouse, such area is completely automated by means of conveyors and automated storage and retrieval machines. The storage area covers a surface of 2000 square meters and is 12 meters high.

Figure 1 shows the whole plant layout of the

manufacturing system.

789

Figure 1: Manufacturing system plant layout

The Assembly workstation represents a bottleneck

in terms of ergonomic issues for the company; in fact the operators always complain of strong back pains at the end of the working shift. A preliminary analysis carried out by the company top management shows that the ergonomic problems are related to the weight of the mechanical parts being moved (mechanical component, flange and assembled workpiece), the wrong design of some working equipments and the objects positions into the workstation. In this regards they asked the authors for designing an improved workstation configuration in terms of acceptable ergonomic risk levels.

Let us describe the operations sequence performed in the Assembly workstation: (1) the operator picks up manually the mechanical component placed on a pallet coming from the Raw materials storehouse and puts it on a first work bench; (2) the operator takes a hand-drill placed on the work bench and uses it for drilling the mechanical components five times; (3) the operator puts away the hand-drill, takes the mechanical component and places it on a second work bench 180 cm far away; (4) the operator takes a flange placed on a pallet coming from the Flange workstation and puts it on the second workbench for performing the assembly operations; (5) the operator puts the flange on the mechanical component and makes them adhering by means of a hammer; (6) the operator fixes together the mechanical component and the flange by using three hexagonal screws located in the screws bin coming from the Screws workstation; (7) the operator gets the assembled workpiece and puts it on a pallet placed 670 cm far away; (8) the operator activates a mechanical arm that gets the assembled workpiece and puts it on a conveyor directed to the Final product warehouse.

3. THE SIMULATION MODEL In this research work the authors present an approach based on the integration of M&S tools and several ergonomics standards (NIOSH81, NIOSH91, Burandt Schultetus, OWAS) for improving the Assembly workstation configuration in terms of interaction between humans and their working environment.

The first step was the development of a simulation model capable of recreating the Assembly workstation with satisfactory accuracy. Before getting into the details of the simulation model development, let us

summarize the three main phases characterizing it: collecting data concerning the Assembly workstation (data collection phase), reproducing the real system in the virtual environment from both a geometric and work method point of view (simulation modelling phase) and verifying if the simulation model is an accurate representation of the real system (validation phase). 3.1. Data collection phase The data collection phase is a mandatory step for developing a simulation model capable of recreating the system under consideration with satisfactory accuracy. In this regards the authors spent a three months period at the Assembly workstation for gathering all the information and data they needed.

Information regarding operator characteristics (age, gender, weight and physical conditions), dimensions (length, height and width), weight and position of all the workstation objects as well as information regarding work method were collected.

The operators characteristics were used for selecting human models capable of representing as much as possible the real operators.

The objects dimensions, weight and position were used for recreating the geometric model of the Assembly workstation.

The work method information were used for reproducing correctly in the virtual environment the workstation assembly process. 3.2. Simulation modelling phase The simulation modeling phase aims at reproducing the Assembly workstation in the virtual environment from both a geometric and work method points of views. This phase consists of two steps:

1. Plant layout generation for recreating the three

dimensional geometric model of the workstation;

2. Human model inserting and training for reproducing in the virtual environment all the operations characterizing the assembly process.

Section 3.2.1 and section 3.2.2 get into the details

of the Simulation Modeling phase main steps. 3.2.1. Plant layout generation The implementation of the geometric models of the Assembly workstation has been made by using the CAD software Pro-E; in this regards, objects dimensions and weight, gathered in the data collection phase, were inserted into the Cad software as input data for designing geometric models with high level of detail. Table 1 reports description, type, dimensions and weight of the objects mainly used during the assembly operations.

790

Table 1: Data collection for geometric models Object

desrciption Object type Dimensions (cm) L x W x H

Weight (Kg)

Mechanical component Component 34,5 x 24,5 x 14 14

Hand drill Equipment 38,5 x 27,5 x 7,5 6

Mechanical arm Equipment 300 x 200 x 100 80

Conveyor Equipment 600 x 100 x 50 55

Screws bin Equipment 30 x 20 15 0,3

Hexagonal wrench Equipment 15 x 5 x 1 0,2

Hammer Equipment 23 x 6 x 2 0,6

Screw Component 4 x 0,5 x 0,5 0,05

Flange Component 29 x 12,5 x 12,5 9

The plant layout generation has been completed by

importing the geometric models into the virtual environment provided by the simulation software eMWorkplace. Note that the geometric models have been located in the same positions the real objects take place in the system in order to recreate exactly in the virtual environment the Assembly workstation.

Figure 2 shows a panoramic view of the virtual assembly workstation.

Figure 2: Virtual Assembly workstation

The side “A” consists of some office furniture and

the mechanical components placed on the pallets coming from the Raw material warehouse.

The side “B” consists of a workbench, a screws bin, a hexagonal wrench, a hammer and flanges; in this part of the workstation the basic assembly operations are performed.

The side “C” consists of several workbenches and hand drills used for drilling the mechanical component.

Finally the side “D” consists of a mechanical arm, a workbench and a conveyor.

Figure 3 and figure 4 show respectively the main mechanical component and the workbench used for the assembly operations.

Figure 3: The mechanical component

Figure 4: Workbench

3.2.2. Human model insertion and training After the plant layout generation, humans models insertion and training are needed for reproducing correctly in the virtual environment all the operations performed in the assembly workstation. As reported in the MANUFACTURING PROCESS section, the assembly workstation employs three operators working each one 8 hours per day on different working shift; on this subject in the simulation model development the authors selected from eMWorkplace libraries a human model capable of reproducing as much as possible the real operators characteristics. In fact all the information concerning age, gender, weight and physical conditions of the three operators were averaged out and used as input data for the human model selection.

The table 2 reports the physical characteristics for each operator and their average values as well.

A

C

B

D

791

Table 2: Operators physical characteristics Operator

Op - 1 Op - 2 Op - 3 averaged Op

Age 41 44 50 45

Gender male male male male

Height (cm) 175 170 178 174

Weight (Kg) 76 69 81 75

Phys

ical

cha

ract

eris

tics

Physical Conditions normal normal normal normal

After the selection of the more suitable human

model, it has been inserted in the virtual environment provided by eMWorkplace.

Obviously, at the beginning, the human model is only able to stand in the waiting position; in fact it needs to be trained for reproducing correctly the assembly operations. To this end, eMWorkplace provides the user with a programming language for teaching the basic motions of each operation such as reach, grasp, realize, move, etc.. In this regard the authors have analyzed each operation trying to identify all the basic motions it is made by. Note that such activity has required a huge amount of time due to the high number of basic motions that generally characterizes an operation.

Hundreds of programming code lines have been written for allowing the human model to recreate exactly the real assembly process. Moreover information regarding working postures at the beginning and end of each lifting task, frequency and duration of lifting tasks have been required by eMWorkplace for evaluating the workstation in terms of ergonomic issues. In effect the software allows the user to analyze the simulation model under consideration by means of several ergonomic standards (NIOSH 81, NIOSH 91, Burandt Schultetus, OWAS); on this subject, after the simulation model runs, the user is provided with the ergonomic standards results so that all the workstation ergonomic problems can be identified..

Figure 5 shows several programming code lines written for teaching the human model.

Figure 5: Some programming code lines

3.3. The validation phase To increase significantly the probability of success in conducting a simulation study, the validation phase is a mandatory step for determining if the simulation model is an accurate representation of the real system under investigation.

The validation phase has been carried out by means of two different approaches: debugging technique and visualization/animation technique.

The debugging technique is an iterative process whose purpose it is to uncover errors or misconceptions that cause the model’s failure and to define and carry out the model changes that correct the errors (Banks,1998). In this regards, during the simulation model development, the authors tried to find the existence of errors (bugs). Given the detected errors, the authors determined the cause of each error and identified the model changes required for their correction. The identified model changes were carried out and further investigations for finding other bugs were accomplished; in effect a change correcting an error may create another one. This iterative process has been continued until no errors were identified in the simulation model.

Visualization/Animation of a simulation model greatly assists the validation phase (Sargent 1996, Bell and O’Keefe 1994). Seeing the animation of the model as it executes and comparing it with the operations of the assembly workstation helped the authors to identify discrepancies between the simulation model and the real system.

Both techniques have been applied with the help of the workstation operators and company production engineers.

The simulation model has been debugged by analyzing all the basic motions of the human model and the virtual workstation has been compared with the real one in terms of both plant layout and work method. In this regard some wrong working postures, wrong motions and redundant motions have been corrected or deleted.

At the end of the validation phase, the authors concluded that the simulation model is capable of recreating the real assembly workstation with satisfactory accuracy.

4. THE ERGONOMIC STANDARDS The main goal of the paper is to achieve the ergonomic effective redesign of the most critical workstation (assembly workstation) of a manufacturing plant by means of an integration between M&S tools and several ergonomic standards. In this regards after the simulation model development, next step was to evaluate the Assembly workstation in terms of ergonomic issues. In effect, eMWorkplace provides the user with several ergonomic indexes based on the most known ergonomic standards (Burandt Schultetus, NIOSH 81, NIOSH 91, OWAS and Garg analysis). The examination of such indexes has allowed the authors to identify all the ergonomic problems related to the workstation. Before

792

getting into the details of the ergonomic analysis, let us describe the ergonomic indexes and the relative ergonomic standards used in the simulation study.

The ergonomic indexes are the lift indexes (evaluated by Lift analysis), the stress levels associated to working postures (evaluated by using the OWAS analysis) and the energy expenditure associated to each activity (evaluated by using the Garg analysis).

The lift analysis consists of three ergonomic methods used for the measurement of the effort that affects the elevation of a weight and its transportation towards another point. The methods are: Burandt Schultetus, NIOSH81 and NIOSH91.

The Burandt-Schultetus analysis detects the maximum weight that a working person can lift (maximum permissible force).

The maximum permissible force can be evaluated by using equation (1):

RFAJCGPF ***= (1)

where, G is a coefficient for the worker’s gender; C is a coefficient for the worker’s health condition; AJ is a coefficient for worker’s age and type of job; RF is the reference force.

Note that the AJ (Age and Job factor) depends on the effort type (i.e. static or dynamic), the worker’s age, the shift time (i.e. 8 hours) and the effort frequency. The RF takes into consideration the torso weight movement, the hands use (i.e. one or two hands), the number of persons performing the operation (i.e. one or two persons), the effect of secondary jobs and the maximum force. In turns, the torso weight movement depends on the lower and upper grasp height and motion frequency; the maximum force depends on body size class (anthropometric measure), upper grasp height and distance of grasp from the body.

Moreover, the method requires several input parameters regarding the physical conditions, age and gender of the worker, the load weight, the lifting frequency (measured in lifts per minute) and the total task duration.

The maximum permissible force is then compared to the current actual force (AF) being exerted. Three different cases can be distinguished:

• Case 1: the maximum permissible force do not exceeds the actual force then an ergonomic intervention is required;

• Case 2: the maximum permissible force is equal to the actual force, then a corrective intervention is necessary in the near future;

• Case 3: the actual force is lower than the maximum permissible force, then no ergonomic intervention is required.

The NIOSH 81 method calculates the action limit

(AL) and the maximum permissible limit (MPL). AL is the weight value which is permissible for 75% of all

female and 99% of all male workers. MPL is the weight value which is permissible for only 1% of all female and 25% of all male workers. Three different cases can be distinguished:

• Case A: the action limit exceeds the maximum permissible limit then an ergonomic intervention is required;

• Case B: the action limit is equal to the maximum permissible limit, then a corrective intervention is necessary in the near future;

• Case C: the actual limit is lower than the maximum permissible limit, then no ergonomic intervention is required.

The NIOSH 91 analysis, additionally to the

NIOSH 81 method, includes the recommended weight limit (RWL) and the lifting index (LI). The RWL is the load that nearly all healthy workers can perform over a substantial period of time for a specific set of task conditions. The LI is calculated as ratio between the real object weight and the recommended weight limit. Three different cases can be distinguished:

• If LI value is less than 1 the lifting task is not

hazardous for some of the population; • If the LI value is equal to 1 the lifting task

could be hazardous for some population; • If the LI value is greater than 1 the lifting task

is hazardous for some of the population.

The LI is the approximate value of the relative level of strain.

Note that NIOSH 81 and NIOSH 91 require as input parameters data regarding worker postures at the origin and destination of the lift, object coupling and the duration of the specified task.

As concerns the OWAS analysis, it carries out a qualitative analysis of the worker's movements during a working process. The analysis calculates the stress associated to each body posture and classifies them in one of the following four stress categories:

• Category 1: the stress level is optimum, no corrective interventions are required;

• Category 2: the stress level is almost acceptable, corrective interventions are necessary in the near future;

• Category 3: the stress level is high, corrective interventions are required as soon as possible;

• Category 4: the stress level is very high, corrective interventions must be carried out immediately.

Finally, the Garg analysis calculates the total

amount of energy spent during the manual operations. The analysis splits up a specified operation into smaller steps calculating for each of them the Energy Expenditure (EE); the sum of these separate steps represents the total Energy Expenditure for the activity.

793

As input parameters, such analysis requires information concerning load weight and body weight as well as gender of the working person.

5. SIMULATION RESULTS In this section the authors discuss the ergonomic analysis results as output of a simulation model run.

The OWAS method was the first one applied in this research work. As previously stated, OWAS is a method that classifies each body posture within one of four stress categories (see section 4). The main advantage of using a simulation tool, as eMWorkplace, is that, during the simulation, the postures are analyzed in real-time and the effects are shown visually. As a result, the program points out which body parts are most affected (figure 3).

Fig. 6: Results of the OWAS analysis

Moreover as soon as the software identifies a

harmful working posture, a message window appears reporting the category it belongs to. Figure 7 shows a message window for a working posture belonging to the Category 3.

Figure 7: Message window for Category 3 working posture When the OWAS method is applied to the Assembly workstation the program assigns a category 3 to the task of taking manually the mechanical component from the pallet. Note that category 3 indicates a high level of effort that produces negative effects on the worker's health, so corrective interventions are required as soon as possible. The color orange, that represents category

3, is visible on the back and the legs of the worker, which are the most affected parts in the movement (figure 8).

Figure 8: Example of a body posture belonging to the category 3

When the operator uses the working tools (hand drill, hexagonal wrench, hammer) for performing the assembly activities, the software indicates a category 2. A category 2 body posture can have adverse effects on the muscular system so that corrective interventions are necessary in the near future. In this case the most affected body part is the back (note that the color yellow is associated to a body part belonging to the category 2).

Finally, the software reports another Category 2 as soon as the worker gets manually the assembled workpiece, whose weight is 23 Kg, and puts it near the mechanical arm location.

As the next step, the ergonomic process is studied through the Lift Analysis.

The Lift Analysis have been used for measuring the strain resulting from human lifting and carrying activities. In this regards, as previously stated in the “ERGONOMIC STANDARDS” section, three ergonomic methods have been carried out: Burandt Schultetus, NIOSh 81 and NIOSH 91.

The first one being applied to the Assembly Workstation was the Burandt Schultetus method. It calculates the maximum permissible force a worker can lift and compare such limit to the actual force being exerted. The input parameters were introduced based on the working cycle, for normal grabbing of the objects and a total duration of the task based on the average time of a Shop Order processing. Note that a typical Shop Order requires the production of 25 assembled workpiece.

Table 3 reports the input parameters inserted into the simulation software for correctly carrying out the Burandt Schultetus analysis.

794

Table 3: Burandt Schultetus input parameters Worker physical characteristics

Physical Condition normal

Age 45

gender male Objects being moved Load weight (Kg)

mechanical component 14

flange 9

hand drill 6

hexagonal wrench 0,2

hammer 0,6

Total task duration (25 workpiece)

Time (sec) 6125

As soon as high stress levels occur, a message

window appears reporting the maximum permissible and actual force values. Moreover the most affected body parts appear orange colored (figure 9).

Figure 9: Results of the Burandt Schultetus analysis When the Burandt Schultetus method is applied to

the Assembly workstation the program detects three critical lifting operations: the grabbing of the mechanical component from the pallet (1), the handling of the hand drill (2) and the carrying of the assembled workpiece to the mechanical arm location (3); in effect each of them is characterized by a PF value lower than the AC being exerted. Table 4 reports for each operation the permissible force and actual force values.

Table 4: Burandt Schultetus returned values

Operation Maximum permissible force (N)

Actual force (N)

(1) 67,8 68,7 (2) 68,3 68,7 (3) 52,3 112,8

As concerns the operations (1) and (2) the PF

values are very close to the AF values; in fact in both cases the AF exceeds the PF no more than 1 N. It means that the stress level is almost acceptable and corrective interventions are necessary in the near future.

On the other hand, the operation (3) has extremely adverse effects on the muscular system; in fact a huge gap between the PF and the AF (the gap value is about 60 N) characterizes it. In this regards, corrective interventions must be carried out immediately.

NIOSH 81 and NIOSH 91 methods complete the Lift analysis. NIOSH 81 calculates the AL and the MPL. NIOSH 91 calculates the RWL and the LI. As input parameters, both methods requires information regarding worker posture at the origin and destination of a lift, object coupling (i.e. grip quality) and the duration of the specified task.

Worker posture at the origin and destination of a lift were identified during the program code writing. In fact, eMWorkplace provides the user with specific commands for denoting the beginning and the end of each lifting operation (in turn, the initial and final worker postures). In this regards the commands “LIFT BEGIN” and “LIFT END” were accomplished respectively just before and after the code lines written for performing the lifting operation.

Grip quality was set as normal for each object being lifted.

Task duration was based on the average time for processing a typical Shop Order made by 25 assembled workpieces; the average time is about 6125 seconds.

NIOSH 81 and NIOSH 91 methods were carried out for each lifting operation performed during the assembly process. As Burandt Schultetus analysis, the grabbing of the mechanical component from the pallet (1), the handling of the hand drill (2) and the carrying of the assembled workpiece to the mechanical arm location (3) were identified as critical movements for the operator.

Table 5 and table 6 show respectively the NIOSH 81 and NIOSH 91 results for the critical operations.

Table 5: Results of NIOSH 81 analysis

NIOSH 81 Operation MPL

(Kg) AL

(Kg) Object

weight (Kg) (1) 31,45 10,48 14 (2) 26,54 8,96 6 (3) 27,58 9,19 23

Table 6: Results of NIOSH 91 analysis

NIOSH 91

Operation RWL (Kg) Object weight (Kg)

LI

Origin 8,83 14 1,58 (1)

Destination 11,30 14 1,23 Origin 6,23 6 0,96

(2) Destination 6,11 6 0,98

Origin 7,26 23 3,16 (3)

Destination 6,15 23 3,73

795

Let us describe the NIOSH 81 results. The MPL values exceed the object weight for each lifting operation. It means that the lifting operations no have adverse effects on the muscular system and, in turn, no corrective interventions are necessary. However, note that the MPL represents the weight value which is permissible for only 1% of all female and 25% of all male workers. We can conclude it is not a significant index due to the little part of the population it takes into consideration. On the other hand, the weight value which is permissible for 75% of all female and 99% of all male workers (action limit) is lower than the objects weight being lifted.

As concerns the operation (2), the AL value is very close to the object weight (the gap is 0,04 Kg) so that the stress level is almost acceptable; in this case corrective interventions are needed in the near future. On the contrary, AL is lower 3,52 Kg and 13,81 Kg than the objects weight respectively for operation (1) and operation (3). In the first case the stress level is high and corrective interventions are necessary as soon as possible, in the second case the stress level is very high and corrective interventions are immediately required.

Concerning the description of NIOSH 91 results, let us consider the LI values. As previously stated, it is calculated as ratio between the object weight and the RWL value.

Operation (1) and operation (3) are characterized by LI values higher than one in both the origin and the destination points of the lifting operations. In the first case the LI values (1,58 and 1,23) suggest corrective interventions as soon as possible. In the second case the very high LI values (3,16 and 3,73) require corrective interventions immediately.

Considering the operation (2), 0,96 and 0,98 represent acceptable LI values so that no corrective interventions are needed.

Finally, the ergonomic study is completed by the Garg analysis. It calculates the total amount of energy spent during the manual operations. Note that such value provides information on the stress rate of the activity. As input parameters it requires information regarding worker physical characteristics (physical condition, age and gender) and the weight of objects being moved (objects weight is reported in table 3). At the end of a simulation run, the total amount of energy spent during the whole working shift is about 1437 Kcal.

Let us summarize the most noteworthy aspects provided by the ergonomic analysis.

The ergonomic process has pointed out the high level of ergonomic risks affecting the Assembly workstation. In particular three operations were identified as the most harmful for the operators muscular system.

The grabbing of the mechanical component from the pallet (1) hurts the back and the legs of the worker.

The handling of the hand drill (2) affects the operator right hand-arm system.

Finally, the carrying of the assembled workpiece to the mechanical arm location (3) causes pain to the worker back and arms.

We can conclude the ergonomic effective design of the workstation is required for preserving the workers health. Note that the ergonomic effective design should also aim at reducing the total amount of energy spent during the assembly process.

Section 6 describes the design guidelines for developing an improved workstation configuration in terms of ergonomic issues.

6. THE ERGONOMIC WORKSTATION

DESIGN Once the initial situation was studied, a new workstation configuration was developed for eliminating the movements that could cause worker injuries.

Let us list the critical operations affecting the workstation and describe for each of them the solution the authors propose for preventing the ergonomic problems.

• the grabbing of the mechanical component

from the pallet: it requires to the worker continuous bending due to the location of the pallet on a hand cart high 20 cm. Operator back and legs are the most affected body parts. The authors decided to substitute the initial hand cart for an adjustable one. This change allows to custom the pallet position in height according to the operators needs. Note that the adjustable hand cart will be used also for moving the assembled workpiece during the assembly process. Figure 10 and figure 11 show respectively the actual configuration and final solution.

Figure 10: Actual workstation configuration for the grabbing the mechanical component from the pallet

796

Figure 11: Final workstation configuration for the grabbing the mechanical component from the pallet

• the handling of the hand drill: it causes pain to the operator right hand-arm system. As previous stated, the hand drill is used for drilling the mechanical component 5 times. In a first moment the authors thought to modify the mechanical component and the hand drill positions for making them easier to manage. Actually, this operation is strongly affected by the hand drill weight (6 Kg), so any change regarding the objects position would have been a useless solution. In this regards, the authors advised the company top management to purchase a lighter hand drill. In particular they proposed a 1.6 Kg hand drill characterized by an ergonomic handle.

• the carrying of the assembled workpiece to the mechanical arm location: 670 cm must be walked carrying manually the assembled workpiece, whose weight is about 23 Kg. Obviously, the worker back and arms are the most stressed body parts. As previous stated, the authors proposes to adopt the adjustable hand cart for performing such operation. In this way, the worker has only to place the assembled workpiece on the hand cart and then push it to the final destination. Figure 12 and figure 13 show respectively the actual configuration and the final solution.

Figure 12: Actual workstation configuration for the carrying the assembled work piece to the mechanical arm location

Figure 13: Final workstation configuration for the carrying the assembled work piece to the mechanical arm location

Figure 14 shows the final workstation

configuration. Two red boxes point out the workstation changes respect to the initial configuration.

Figure 12: Final Assembly workstation

797

The use of the simulation allows to check the goodness of the changes before they are implemented in the real system. To this end, the final workstation configuration has been tested by means of the simulation model. In this regards the ergonomic analysis have been carried out for evaluating the ergonomic risks inside the proposed workstation configuratiom. The simulation results point out the ergonomic effective design of the workstation. In particular the ergonomic issues related to the three critical operations have been solved and no other ergonomic problems have been detected.

The OWAS analysis does not reveal any particular posture problem. The stress level related to each body posture is optimum.

According to the Lift analysis (Burandt Schiltetus, NIOSH 81 and NIOSH 91) no lifting problems affect the workstation.

Consider the Garg analysis. The total amount of energy spent during the whole working shift is about 1203 Kcal. It means that the EE reduction is about 17% respect to the initial situation.

In conclusion, the goodness of the configuration is also sustained by the low economic efforts required for the workstation changes implementation in the real system. In effect the company has only to incur the costs related to the purchase of the hand drill and the adjustable hand cart. Table 7 reports the objects costs.

Table 7: Objects costs

Object Costs ($)

Adjustable hand cart 594.40

Hand drill 60.00

Total 654.40 The workstation relevant ergonomic improvements

completely justified the total amount of 654.40 $.

7. CONCLUSION The main goal of the paper is to achieve the ergonomic effective redesign of the most critical workstation (assembly workstation) of a manufacturing plant by means of an integration between M&S tools and several ergonomic standards. The authors started the research work by modelling the actual configuration of the workstation. The simulation model has been developed by using the CAD software Pro-Engineer and the simulation software eMWorkplace. After the simulation model validation, several egonomic analysis were accomplished for identifying the ergonomic problems related to the actual workstation. A new workstation configuration was developed for eliminating the movements that could cause worker injuries.

The final configuration of the workstation has acceptable ergonomic risks (related to lifting activities and working postures) and guarantees a smaller amount of energy required for performing all the operations.

Further researches are still on going (in cooperation with the same company) for analyzing the remaining workstations of the Production area. REFERENCE Banks. J., 1998. Principles of Simulation, Handbook of

Simulation. New York: Wiley Interscience. Bell, P.C., O’Keefe, R.M., 1994. Visual interactive

simulation: a methodological perspective. Annals of Operations Research, 53, 321-342.

Bocca, E., Longo, F., 2008. Simulation Tools, Ergonomics Principles and Work Measurement Techniques for Workstations Design. Proceedings of Summer Computer Simulation Conference, 15-18 June, Edinburgh (UK).

Carrasco, C., Coleman, N., Healey, S., 1995. Packing products for customers, An ergonomics evaluation of three supermarkets checkouts. Applied Ergonomics, 26, 101-108.

Chang, S.-W., Wang, M.-J. J., 2007. Digital Human Modeling and Workplace Evaluation: Using an Automobile Assembly Task as an Example. Human Factors and Ergonomics in Manufacturing, 17, 445-445.

Das, B., Sengupta, A. K., 1996. Industrial workstation design: a systematic ergonomics approach. Applied Ergonomics, 27, 157-163.

De Sensi, G., Longo, F., Mirabelli, G., 2007-a. Modeling & Simulation Based Approach for Optimizing Seal Press Workstation in a Manufacturing System, Proceedings of Business and Industry Symposium, March 25-29, USA.

De Sensi, G., Longo. F., Mirabelli, G., 2007-b. Ergonomic work methods optimization in a three dimensional environment, Proceedings of Summer Computer Simulation Conference, July 15-18, San Diego, California, USA

Engström, T., Medbo, P., 1997. Data collection and analysis of manual work using video recording and personal computer techniques. International Journal of Industrial Ergonomics, 19, 291-298.

Feyen, R., Liu, Y., Cha, D., Jimmerson, G., Joseph, B., 2000. Computer-aided ergonomics: a case study of incorporating ergonomics analyses into workplace design. Applied Ergonomics, 31, 291-300.

Garg A., 1976. A metabolic rate prediction for manual materials handling jobs. Dissertation, University of Michigan.

Grant, K.A., Habes, D.J., Bertsche, P.K., 1997. Lifting hazards at a cabinet manufacturing company: Evaluation and recommended controls. Applied Occupational and Environmental Hygiene, 12, 253-258.

Jones, T., Strickfaden, M., Kumar, S., 2005. Physical demands analysis of occupational tasks in neighborhood pubs. Applied Ergonomics, 36, 535-545.

Kadefors, R., Forsman, M., 2000. Ergonomic evaluation of complex work: a participative approach employing video computer interaction,

798

exemplified in a study of order picking. International Journal of Industrial Ergonomics, 25, 435-445.

Kharu, O., Harkonen, R., Sorvali, P., Vepsalainen, P., 1981. Observing working postures in industry: Examples of OWAS application. Applied Ergonomics, 12, 13-17.

Lin, R.T., Chan, C.-C., 2007. Effectiveness of workstation design on reducing musculoskeletal risk factors and symptoms among semiconductor fabrication room workers. International Journal of Industrial Ergonomics, 37, 35-42.

Longo, F., Mirabelli, G., Papoff, E., 2006. Effective Design of an Assembly Line Using Modeling & Simulation. Proceedings of the Winter Simulation Conference, Monterey, California, USA.

Longo, F., Mirabelli, G., Papoff, E., 2005. Techiche di analisi avanzate per la progettazione efficiente delle postazioni di assemblaggio manuale, SdA – Soluzioni di Assemblaggio, VNU Business Publications Italia.

Longo, F., Mirabelli, G., Papoff, E., 2006-a. Material Flow Analysis and Plant Lay-Out Optimization of a Manufacturing System, International Journal of Computing, 5(1), 107-116.

McAtamney, L., Corlett, E.N., 1993. RULA: a survey method for the investigation of work-related upper limb disorders. Applied Ergonomics, 24, 91-99.

Mital, A., Ramakrishnan, A., 1999. A comparison of literature-based design recommendations and experimental capability data for a complex manual materials handling activity. International Journal of Industrial Ergonomics, 24, 73-80.

Neumann, W.P., Wells, R.P., Norman, R.W., Kerr, M.S., Frank, J., Shannon, H.S., OUBPS Working Group, 2001. Trunk posture: reliability, accuracy, and risk estimates for low back pain from a video based assessment method. International Journal of Industrial Ergonomics, 28, 355-365.

Niosh Technical Report 81-122. National Institute for Occupational Safety and Health (Hrsg.). Work practices guide for manual lifting. Center for Disease Control, U.S. Department of health and human services, Cincinnati, OH, USA: NTIS 1981.

Russell S. J., Winnemuller L., Camp J. E., Johnson P. W., 2007. Comparing the results of five lifting analysis tools. Applied Ergonomics, 38, 91-97.

Russell S. J., Winnemuller L., Camp J. E., Johnson P. W., 2007. Comparing the results of five lifting analysis tools. Applied Ergonomics, 38, 91-97.

Santos, J., Sarriegi, J. M., Serrano, N., Torres, J. M., 2007. Using ergonomic software in non-repetitive manufacturing processes: A case study. International Journal of Industrial Ergonomics, 37, 267-275.

Sargent, R.G., 1996. Verifying and validating simulation models. Proceedings of the Winter Simulation Conference. pp. 55-64. Piscataway (New York, USA).

Schultetus, W., 1980. Daten, hinweise und beispiele zur ergonomischen arbeitsgestaltung. Montagegestaltung, Verlag TÜV Rheinland Gmbh, Köln.

Scientific Support Documentation for the Revised 1991 NIOSH Lifting Equation: Technical Contract Reports, May 8, 1991, NTIS No. PB-91-226-274.

Scott, G. B., Lambe, N. R., 1996. Working practices in a pherchery system, using the OVAKO Working posture analysing System (OWAS). Appied Ergonomics, 27, 281-284.

Temple, R., Adams, T., 2000. Ergonomic Analysis of a Multi-Task Industrial Lifting Station Using the NIOSH Method. The Journal of Industrial Technology, 16, 1-6.

Van Wendel de Joode, B., Burdorf, A., Verspuy, C., 1997. Physical load in ship maintenance: Hazard evaluation by means of a workplace survey. Applied Ergonomics, 28, 213-219.

Vedder, J., Hellweg, 1998. Identifying postural hazards with a video-based occurrence sampling method. International Journal of Industrial Ergonomics, 22, 4-5.

Waters, T.R., Vern, P.A., Garg, A., 1994. Application Manuals for the Revised NIOSH Lifting Equation. U.S. Department of health and human services, National Institute for Occupational Safety and Health, Cincinnati, OH, USA.

Wright E.J., Haslam R.A., 1999. Manual handling risks and controls in a soft drinks distribution centre. Applied Ergonomics, 30 , 311-318.

AUTHORS BIOGRAPHY ANTONIO CIMINO was born in Catanzaro (Italy) in October the 1th, 1983. He took his degree in Management Engineering, summa cum Laude, in September 2007 from the University of Calabria. He is currently PhD student at the Mechanical Department of University of Calabria. His research activities concern the integration of ergonomic standards, work measurement techniques, artificial intelligence techniques and Modeling & Simulation tools for the effective workplace design. He collaborates with the Industrial Engineering Section of the University of Calabria to research projects for supporting innovation technology in SMEs.

GIOVANNI MIRABELLI was born in Rende in 1963 and he took the degree in Industrial Engineering at the University of Calabria. He is currently researcher at the Mechanical Department of University of Calabria. His research interests include ergonomics, methods and time measurement in manufacturing systems, production systems maintenance and reliability, quality. He has published several scientific papers participating as speaker to international and national conferences. He is actively involved in different research projects with Italian and foreign universities as well as with Italian small and medium enterprises.

799