Acceptance sampling plan has been a practical tool in quality management, which provides both the producer and the consumer a general rule for sentencing lots. In recent years, a new strategy, chain sampling plan, considering disposition of historical lots for sentencing the current lot has been discussed. However, reviewing results of preceding lots only happens when the quality of current lot is on the undefined region, which may not be completely utilized. Therefore, in this study, we develop a variables modified‐ChSP (VMChSP) for variable inspection based on the Taguchi capability index to improve above situation. Optimization model of the proposed plan is formulated to solve plan parameters by minimizing the required sample size. Plan parameters under various allowable risks and quality levels are also provided for practical purpose. Finally, further analysis and comparisons between the proposed plan and conventional plans are provided and discussed in this research.