Considering the disadvantages of the existing Joint Replenishment Problem (JRP) model under real production system, we discuss a JRP model with budget and storage capacities restrictions. As a typical NPhard problem, there is no stable and effective algorithm to deal with the proposed JRP model. This paper presents a reliable adaptive hybrid differential evolution (AHDE) algorithm which is applicable to the JRP with resource restrictions based on the test with typical functions. Most of the research work was done under the assumptions that the demand for each item type, minor ordering cost and the unit holding cost are known and constant. However, the real values of these parameters may change over the replenishment horizon and are hard to be estimated precisely in the real world. The present study addresses the sensitivity analysis and discusses how misestimates of these parameters may influence the replenishment policy as determined by the proposed method.