Traditional double variable sampling plan is an inspection with crisp parameter, so that it is not able to make a decision about manufacturing products whose proportion parameter ( p) is uncertain. The major purpose of this study is to design double variable sampling plan when p is imprecise to inspect the lots of manufacturing products as interested characteristics are normally distributed. The plan parameters are obtained by optimization problem. During the process of solving optimization problem, sum of the fuzzy customer and producer's risks and contract's requirements are considered as objective function and constraints, respectively. The optimal parameters are prepared to use in industrial environments for various combinations of requirements. We indicate that the proposed scheme turns to classical one as p is not vague. The obtained results presents that the introduced scheme is more economical than the available plan. Finally, a practical example is presented in real applications.