Cell staffing and standardized work design in Chaku-Chaku production lines using a hybrid optimization algorithm

• Published in: 2009 International Conference on Computers and Industrial Engineering pp. 305 - 310
• Main Authors: Wenming Zhang, Deuse, J.
• Format: Conference Proceeding
• Language: English
• Published: IEEE 01.07.2009
• Subjects: Algorithm design and analysis; Assembly; Chaku-Chaku production lines; Constraint optimization; Design optimization; Gantt chart; Integer linear programming; Manufacturing industries; MILP; MILP/CP hybrid model; Mixed integer linear programming; Production; Resource management; Standardization; standardized work
• ISBN: 9781424441358; 1424441358
• DOI: 10.1109/ICCIE.2009.5223901

Chaku-Chaku production lines are currently widely adopted in manufacturing and assembly industry, which are an effective and efficient production organizational type thanks to its flexibility as well as customer-oriented flow layouts. However, due to the variation of customers' demand and unpredictable market factors, the layout as well as manpower allocation are both facing to be adjusted to satisfy the customers' orders, compared with reconstruction of machines or redesign of new production line, the adjusted focus is shifting to the workforce allocation and work reassignment, for the utilization of bottleneck resource machines can achieve at most 50%-60%. This work specifies the calculation of number of the operators employed to run the production line and work assignment for each operator, i.e. standardization of the production routines and the design of optimal standardized work for every operator. In order to simplify such combinatorial optimization problem, the number of manpower and the detailed work assignment are separated to be stated by mixed integer linear programming model (MILP) and constraint programming model (CP), the first model is used to optimize the workforce allocation subject to several internal constraints, the following model illustrates how to achieve the detailed scheduling, which is utilized to assign detailed work to each operator and further standardize the work sequence and content. MILP is constructed without consideration of each feasible operational sequence as well as cooperation between adjacent workers so as to simplify the assignment problem and then realize the calculation of optimal workforce. Combined with CP model as well as with the help of feedback loops, the optimal and feasible standardized work can be finally depicted in the form of Gantt chart. The performance of the approach is also verified through comparison with the stated approach in the currently available literature.