Abstarct: Testing statistical hypotheses are one of the most commonly used statistical issues in decision making. In the usual methods of testing statistical hypotheses, data, hypotheses, parameters, and other precise problem elements ‎are accurate‎. But in applied sciences, such as economics, agriculture, and the social sciences, we may encounter ambiguous definitions and fuzzy concepts. In such a situation, classical methods require generalization in fuzzy environments. Entry of ambiguity in the hypothesis testing problem can be done through data or hypotheses. Therefore, three major issues can be considered 1. Exact hypothesis testing with inaccurate (fuzzy) data, 2. Examination of inaccurate hypotheses with accurate data, 3. Examination of non-accurate hypotheses with inaccurate data.‎‎ One of the most commonly used methods for hypothesis testing is the p-value approach. In this thesis, we implement hypothesis testing in fuzzy environments baxsed on p-value and show how to calculate p-value in R software with two applications packages Fuzzy p-value and FPV. In addition, we make a balanced hypothesis test baxsed on the comparison of the fuzzy p-value test of the hypothesis ‎H0‎ versus ‎‎H‎1‎‎ and the p-value of the fuzzy test ‎H1‎‎ versus ‎‎H0‎‎ in a fuzzy environment. We also describe an applied approach with regard to fuzzy assumptions in quality of control.