Development of simulation models for the cement loading process at a cement plant

Srisurin, Punyaanek
Journal Title
Journal ISSN
Volume Title
[Honolulu] : [University of Hawaii at Manoa], [December 2013]
In queuing systems, waiting time and idle time are considered nonproductive since waiting time limits opportunities to improve productivity, while idle time creates wasteful operation costs. Bulk cement loading process is a good example to illustrate the problem in logistics distribution system. Reduction of waiting time and throughput time of cement trucks in cement loading process requires such additional resources as loading machines and operators, which trade off additional operation costs for cement distributors; however, an abundant number of input resources could create idle time. On the other hand, minimizing the input resources can significantly save costs for entrepreneurs; however, waiting time and throughput time of trucks could be excessive that limits the number of trips that trucks could travel to transport cement. To maximize utilities of cement trucks and input resources, a trade-off between truck waiting time and server idle time needs to be resolved. The current practice yields average waiting time of 26.0 minutes per truck and throughput time of 71.5 minutes per truck. According to the survey conducted by the plant, the current practice reflects the declining in customer satisfaction since the throughput time is high. To say, the current practice causes, on average, 47.5 minutes of non-productive time per truck, compared to 24.0 minutes of cement loading time. Consequently, simulations need to be performed for the sake of determining some balance between truck waiting time and idle time of servers. The primary objective of this study is to develop a simulation model that maximizes the utilization factor of truck waiting time and machine idle time in cement loading process. In addition, among the models showing maximum utilization factor, the study aims to pick up the potential model that can reduce truck waiting time and queue length, with as lowest machine idle time as possible. By using EZStrobe software, the current approach of the cement loading process was simulated based on the data obtained between August 2012 and June 2013. Model validation was performed by comparing results from the process simulation model with the results obtained from statistics to indicate accuracy of the simulation approach before proceeding to making improvements on the model. Various simulation models were developed by adjusting the process and allocating resources to predict total costs, average idle time of loading machines, average waiting time, average queue length, and average throughput time of trucks for each model. Sensitivity analysis were performed to determine the best practice that yields the greatest utilization factor and improve the overall performance of the process by minimizing truck waiting time, queue length, and machine idle time, with reasonable corresponding cost. The research was conducted based on a case study of the cement loading system at Asia Cement Plant by using population data collected during August 2012-June 2013. The current model and improved simulation models, alongside with their results, were graphically and numerically presented to illustrate potential of the models in terms of maximizing the utilization factor. In addition, in-depth analyses were performed on the raw data to determine appropriate statistical distributions and characteristics for all parameters used in the simulation models.
M.S. University of Hawaii at Manoa 2013.
Includes bibliographical references.
Civil engineering
Access Rights
Email if you need this content in ADA-compliant format.