Acceptance sampling is a useful tool for determining whether submitted lots should be accepted or rejected. With the current increase in outsourcing production processes and the high-quality levels required, it is very desirable to have an efficient and economic sampling scheme. This paper develops a variables repetitive group sampling (RGS) plan that accounts for the process yield (meeting the manufacturing specifications) and the quality loss (variation from the target). The plan parameters are determined by solving a nonlinear optimisation problem. This implies that the plan parameters minimise the average sample number required for inspection and fulfil the classical two-point conditions on the operating characteristic (OC) curve. Besides, this paper investigates the efficiency of the proposed plan and compares it with the existing variables single sampling plan. Tables of the plan parameters for the proposed variables RGS plan are provided and an application example is presented for illustration. IntroductionAcceptance sampling is one of the most practical tools in classical quality control and assurance applications, which deal with quality contracts for product orders between factories and their customers. Acceptance sampling plans provide the producer and the consumer with a general criterion for lot sentencing. A well-designed sampling plan can substantially reduce the difference between the required and the actual supplied product quality Wu 2006, 2007). Unfortunately, it cannot avoid the risk of accepting unwanted poor product lots, nor can it avoid the risk of rejecting good product lots without implementing 100% inspection (e.g. Montgomery 2009). The criteria used to measure the performance in an acceptance sampling plan are usually based on the operating characteristic (OC) curve, which quantifies the risks of producers and consumers. The OC curve plots the probability of accepting a lot against the actual quality level of the submitted lots. In other words, the OC curve shows the discriminatory power of the sampling plan, which provides the producer and the buyer with a common base for judging whether the sampling plan is appropriate. Sherman (1965) developed a new type of sampling plan, called the repetitive group sampling (RGS) plan, for attributes. The operating procedure of this RGS plan is similar to that of the sequential sampling plan. Balamurali and Jun (2006) extended the RGS concept to variables inspection for a normally distributed quality characteristic. They also compared the efficiency of the variables RGS plan with the variables single and double sampling plans. Their results indicate that the variables RGS plan gives the desired protection with the minimum average sample number (ASN) for inspection. It is highly desirable to have an efficient and economic acceptance sampling scheme, especially when the required quality level is very high (Montgomery 2009). Therefore, the main purpose of this paper is to develop a variables RGS plan for product acceptance determination based on the ...

This research develops a heuristic algorithm for assembly line balancing problem (ALBP) of stitching lines in footwear industry. The proposed algorithm can help to design the stitching line with workstations, machines and operators for the production of every new product model. Rank-positional-weighted heuristics and hybrid genetic algorithms are proposed to solve ALBP. First, the heuristics assign tasks and machines to workstations. This solution is then used as an initiative population for hybrid genetic algorithm for further improvement. Real data from footwear manufacturers and experimental designs are used to verify the performance of the proposed algorithm, comparing with one existing bidirectional heuristic. Results indicate that when the size and shape of shoes increase, the proposed genetic algorithm achieves better solution quality than existing heuristics.Production managers can use the research results to quickly design stitching lines for short production cycle time and high labor utilization.