PENGEMBANGAN METODA PENENTUAN DUE DATE ORDER TOOLS BONDING COMPOSITE DI PT IPTN

• Main Author: Abror (NIM : 23494026), Nadjamuddin
• Format: Theses
• Language: Indonesia
• Online Access: https://digilib.itb.ac.id/gdl/view/4707
• Institution: Institut Teknologi Bandung

<b>Abstract :</b><p align=\"justify\"> <br /> <br /> The ability of tools shop to deliver tool manufacturing order on time according to the need schedule of the requiring part is very important. The reason is that failure to complete tools on time may cause slippage to aircraft production schedule, additional tangible costs (e.g. clerical work, plant overtime, inventory), as well as intangible costs (e.g. loss of customer satisfaction). <br /> <p align=\"justify\"> <br /> This thesis concerns with the development of procedure for simplifying actual complex shop using the Optimized Production Technology (OPT) approach. In addition to the shop modeling, a mathematical model which combines product characteristics such as routing and processing times with external factors such as shop loads and shop capacity is also proposed. In the development of the mathematical model, the generic queuing system G/G/m/oo is used in combination with the First Come First Served (FCFS) priority discipline. Using the queuing theory principles, total waiting time in the shop is formulated as a multiple factor k of total processing time of the job. The proposed queuing model is then applied to the simplified shop model. <br /> <p align=\"justify\"> <br /> A computer program in Turbo Pascal is developed to support implementation of the mathematical model in daily activity. The program requires input data such as the tools family or the job type including their product structures and standard routings, historical data of interarrival time for each job type, and the workcenters capacity. <br /> <p align=\"justify\"> <br /> The model is validated using hypothetical data. A simulation in GPSS/H is generated on the hypothetical shop to represent the actual condition. The mathematical model performance is measured by AK defined as a ratio of the difference between k calculated using mathematical model and k derived from simulation to the k derived from simulation. For the shop with utilization below 0.837, the model gives AK of 4.47 %.