Implementation of hybrid dynamic work allocation methods

Labor turnover can be very detrimental to the performance of traditional serial assembly lines (TAL). It has been shown that high turnover rates can reduce annual production throughput in serial assembly lines by more than 16% when these lines operate under a balanced work allocation strategy. In TAL the work content allocation is usually performed by equally dividing the work content among each operator in the line. This work allocation assumes equal processing speed among the operators to minimize blocking and starvation in the line, but maintaining an equal speed becomes a difficult task in the presence of labor turnover. The learning process that a new operator must undergo when first introduced to an assembly line station generates a differential in the speeds of the operators, reducing the throughput of the line. An alternative to TAL is dynamic work allocation models (DWA) such as bucket brigades (BB) and work sharing methods (WS), which have shown to work better in the presence of high labor turnover. In this presentation we show a new hybrid method called modified work sharing method (MWS), which combines the best characteristics of TAL, BB and WS designs. This new method has the objective of improving the throughput of serial assembly lines in highly variable environments. Although MWS offers a better alternative to TAL in the presence of labor turnover, it could be more complicated to manage. Then the purpose of the present research is to provide the tools to manage and improve the operation MWS assembly lines. In this presentation we focus on the design guidelines required for the construction, implementation and operation of a MWS assembly line in an industrial setting, such as the number of workstations, buffer size and operational rules. Particular emphasis will be given to the operational rules that are involved in controlling and modifying the assembly line as the conditions in the system change through time, to allow the line supervisor to make decisions regarding the timing and the methodology to make changes to the existing assembly line design. This is based on the observation that in an assembly line the supervisor should exercise pro-active changes to the operational parameters of the line to optimize its performance. Some of these actions include: the reallocation of workers as they gain experience and assembly speed, and also a modification of the level of worksharing by changing the number of parts in the control buffers. Finally we will talk about of some of the tools that supervisors require for assessing the conditions of the line, such as visual performance rules that allow the supervisor to select adequate changes into the assembly line.