Per Medbo

A case study evaluating the ergonomic and productivity impacts of partial automation strategies in the electronics industry

A case study is presented that evaluates the impact of partial automation strategies on productivity and ergonomics. A company partly automated its assembly and transportation functions while moving from a parallel-batch to a serial line-based production system. Data obtained from company records and key informants were combined with detailed video analysis, biomechanical modelling data and field observations of the system. The new line system was observed to have 51% higher production volumes with 21% less per product labour input and lower work-in-process levels than the old batch-cart system. Partial automation of assembly operations was seen to reduce the total repetitive assembly work at the system level by 34%. Automation of transportation reduced transport labour by 63%. The strategic decision to implement line-transportation was found to increase movement repetitiveness for operators at manual assembly stations, even though workstations were constructed with consideration to ergonomics. Average shoulder elevation at these stations increased 30% and average shoulder moment increased 14%. It is concluded that strategic decisions made by designers and managers early in the production system design phase have considerable impact on ergonomic conditions in the resulting system. Automation of transport and assembly both lead to increased productivity, but only elements related to the automatic line system also increased mechanical loads on operators and hence increased the risk for work-related disorders. Suggestions for integrating the consideration of ergonomics into production system design are made.

Data collection and analysis of manual work using video recording and personal computer techniques

Abstract This paper initially reports on data collection and analysis of assembly systems including manual work using the so-called ‘loss analysis’ which we have applied within the automotive industry during two decades of research and development work. In the further development of this method of analysis we have during the last few years synchronized a personal computer and a video recorder, thereby facilitating data collection of true shop floor information, as well as developing the analysis procedure to also include ergonomic aspects and work performance in the analysis. This paper reports on: (1) the background, (2) the prototype equipment used for data collection and analysis which is illustrated by empirical data using the equipment to analyse stamping of sheet metal, as well as assembly of automobile doors and drivelines, and (3) how to integrate a ‘loss analysis’ with an ergonomic analysis of working postures (in this specific case using the so-called OWAS technique) studying assembly work on an automobile door. In this case we noted That the division of labour loss represents over 60% of the time consumption for assembly work. This analysis indicates that from an ergonomic point of view it is primarily worth studying what activities are included in the assembly work classified as OWAS work load level 2. Relevance to industry This paper describes and illustrates methods and equipment developed consisting of a synchronized personal computer and video recorder. This promotes a design based on empirical data for specific cases through facilitating the collection and analysis of shop floor information about manual work and process efficiency of interest for the design of assembly systems.

Integrating human factors into discrete event simulations of parallel flow strategies.

This article demonstrates an approach to integrating human factors (HF) into a discrete event simulation (DES) study of semi-parallelised production strategies. Operationalised HF included operators’ autonomy at work, a known workplace health factor; and reduced operator capability, a factor for new or injured operators. These HF were tested in scenarios of serial flow and two degrees of semi-parallelised flow. The parallel production systems demonstrated better productivity than serial flow in all conditions tested and were less affected by either of the HF tested. It is concluded that HF can be integrated into DES which facilitates early consideration of operator risk and system vulnerabilities. Parallel production approaches, although less common in practice, appear to have advantages over serial flow in terms of productivity, injury risk and the accommodation of operators with temporarily reduced capacity. Further research should expand the range and test the validity of HF integrated into DES modelling procedures.

Assembly station design: A quantitative comparison of the effects of kitting and continuous supply

Purpose – The purpose of this paper is to determine how kitting, compared to continuous supply, affects the time spent by the assembler fetching parts in manual assembly. Design/methodology/approach – The paper is based on an experimental setup at the Saab Automobile assembly plant in Trollhattan, Sweden. Experienced assemblers were studied as they performed the same assembly operations in ten different configurations. Each configuration consisted of a different arrangement in terms of how parts were presented. The use of kits to present parts was compared to parts presentation through continuous supply, where each part number was presented in a separate container. Findings – The time for fetching parts is significantly shorter when parts are presented through kitting instead of through continuous supply. Furthermore, the shorter fetching time is not just related to a shorter distance between assembly object and parts presentation, which can often be achieved through kitting. The reduction of time spent searching for parts is also considerable. Practical implications – The results of the paper provide valuable input in the design of assembly and materials supply systems, as they enable a better understanding of the relative performance of the materials feeding principles of kitting and continuous supply. Originality/value – Previous studies of kitting and its impact on assembly are mostly conceptual or qualitative, whereas quantitative studies are scarce. The current paper provides a substantial contribution by quantifying the effects that kitting, compared to continuous supply, has on the time spent fetching parts.

Simulating operator learning during production ramp-up in parallel vs. serial flow production

The aim of this research is to demonstrate how human learning models can be integrated into discrete event simulation to examine ramp-up time differences between serial and parallel flow production strategies. The experimental model examined three levels of learning rate and minimum cycle times. Results show that while the parallel flow system had longer ramp-up times than serial flow systems, they also had higher maximum throughput capacity. As a result, the parallel flow system frequently outperformed lines within the first weeks of operation. There is a critical lack of empirical evidence or methods that would allow designers to accurately determine what the critical learning paramters might be in their specific operations, and further research is needed to create predictive tools in this important area.

Utilising the potential of combining local and global supply chains

Outsourcing of manufacturing leaves two options for the structuring of the supply network: either complete outsourcing of an item or process, or combining external sourcing and in-house manufacturing. The objective of this paper is to explain the effect on performance of various strategies of combining in-house manufacturing and outsourced manufacturing. The analysis is based on a case study of a European manufacturer (EM) that has outsourced part of its casting processes to Chinese suppliers. The strategies were tested by means of a simulation study and compared in terms of their effects on inventory performance. The study showed at significance level p<0.01 that inventory levels in China decreased 23% and inventory levels in Europe decreased 6% with a so-called base-surge strategy, sourcing predictable and stable demand in China and uncertain demand in-house in Europe.

Inter-relations between product variant codifications and assembly work for flexible manufacturing in autonomous groups

This paper reports on the inter-relations between product variant codification used for specifying product variants and assembly work on the shop floor within a Swedish automobile corporation. The authors give an overview of a general codification scheme as well as product variants and also show how product variants are described from an assembly point of view in two final assembly plants. One plant has an old traditional short cycle-time assembly line, while the other new plant has a long cycle-time, parallelized flow and small work groups. Finally, the authors analyse how the product variants influence the assembly work. To conclude, a reformation of the product variant description is performed, which implies a far less complex product from an assembly point of view than implied by the traditional description methods.

Integrating Production Engineering and Ergonomics in Production System Design

Within the framework of the COPE program, a tool is described for integrated documentation and prediction of ergonomic and technical performance in production systems. The tool is based on data on exposures and durations of tasks occurring in production. A case study is reviewed to illustrate initial efforts to implement the tool, as well as further lines of its development.

Representative video recordings for synchronized collection of ergonomic and production data in an assembly system

A method has been developed for optimized representative video recordings for synchronized collection of ergonomic and engineering data in assembly systems using the same film. The method is based on random selection of both operators and starting time of the recordings during the day with some boundary conditions: fixed total recording time (30h) and an upper (3h) and lower (1h) recording time per operator. The method was tested at an electronic assembly industry. The study showed that the method has to be further developed. This development would be on basis of statistical models assessing associations between measurement resources, representativeness and precision as functions of working hours, number of work stations, manning and work rotation.