Companies worldwide try to employ contemporary manufacturing systems that can cope with changes in external competitive environments and internal process variability. Just In Time (JIT) philosophy helps achieve the required resilience by its policy of having people, machines, and material just-in-time for any given process. U-shaped assembly lines (U-lines) are used to implement JIT principles. Another principle that helps achieve competitive advantage by developing a flexible workforce that responds efficiently to change is that of work-sharing. Operators share work and help each other in a dynamic and floating way, requiring little management effort to distribute workload amongst operators, or balance the assembly line.

The aim of this work is to develop an effective work-sharing protocol for U-shaped assembly lines that will provide the combined advantages of U-lines and work-sharing principles. The new protocol is based on two ideas from literature - the Cellular Bucket Brigade (CBB) system, and the Modified Work-Sharing (MWS) system. To keep the focus on developing the protocol, the scope of this work was limited to two worker systems. The methodology used is to model the protocol and U-line system as a discrete event simulation model, and then use an optimization model to maximize throughput and find optimal buffer locations and levels. A physical simulation experiment was conducted in the Toyota Production Systems lab at RIT to validate the model. Once validated, computer simulation experiments were run with industry data, and results obtained were compared with existing protocols from literature.

It was found that the new protocol performed at least as well as the CBB protocol, improving the output by an average of 1%, for the scenarios tested. Increase in processing speed variability as well as larger variation among workers were found to negatively impact the performance of the protocol. The results were analyzed further to understand why these factors are significant, and why there are anomalies and patterns, or lack thereof. Finally, limitations of the protocol, and opportunities for future research in the field are presented. Major limitations of the protocol are that it is difficult to comprehend, and the assumption of an assembly line divided into equal tasks is not practical in the industry.

Publication Date


Document Type


Student Type


Degree Name

Industrial and Systems Engineering (MS)

Department, Program, or Center

Industrial and Systems Engineering (KGCOE)


Andres L. Carrano


Physical copy available from RIT's Wallace Library at TS178.4 .S74 2013


RIT – Main Campus

Plan Codes