Purpose: Existing productivity improvements activities such as inventory buffer, overall equipment effectiveness (OEE) and total productive maintenance (TPM) do not analytically associate the throughput shortage with process parameters. The paper aims to develop and test an integrated model to recover the throughput shortage through process parameters adjustment in a semiconductor assembly. Design/methodology/approach: The mathematical model of planned throughput is developed as a function of input parameters in an integrated multiple process line. When the planned throughput does not meet the real-time throughput, the throughput shortage occurs. The planned throughput for the next day is summed with the throughput shortage from previous day and mathematical programming is used to find the optimum process parameter values. Findings: The throughput shortage is restored using mathematical programming for the subsequent day of planning. If there still exist throughput shortage, the additional throughput shortage will be carried forward for the subsequent day of planning where mathematical programming is used again to find the adjusted process parameters. The proposed optimization model in essentially a parametric model using input parameters of real-time data based on a normal distribution that is translated into a range of minimum and maximum using 95% confidence interval. Research limitations/implications: The adjusted input parameters in this model is based on the cycle time of three processes which are Die Attach, Wire Bond and Pre-Cap Inspection. Downtime and setup time are not subjected to adjustments. Practical implications: The mathematical programming computes optimum process parameters values to restore the throughput shortage where it correlates the process parameters and throughput shortage quantitatively rather than conventional method of throughput shortage recovery. Originality/value: The research addresses the process parameter adjustment to recover the throughput shortage in integrated multiple processes.
