In recent years, the level of competition extremely increased for all markets. Due to globalization, similar products and services are offered by many players. In order to become stronger in this competitive environment, all players must act strategically and enhance their supply chain management activities. Some companies try to extend their market share from wholesale level to retail and lower levels. This strategy forces companies to reach customers by vendor machines. The aim of this study is to develop a mathematical model to optimally manage operations of vendor machines carrying multi-items located at different locations. The considered problem handles pricing, routing, inventory management and capacity planning problems. We seek an optimal solution that will answer the following research questions: (1) Which items should be available at each location and how many towers should be allocated for those items? (2) What should be the optimum inventory level of these items at the beginning of each period? (3) Shall there be any item transfer between locations at each period of time? (4) What should the price for each item at each location and at each time period be? Our model considers the uncertainty in demand of the items with the help of scenarios. The proposed model is checked by several functional tests to verify that the model works properly and gives the expected results. By conducting numerical experiments, we understand that our model provides efficient solutions to the problem with uncertainties. We also compare the results of the model with the results of a heuristic approach. When compared with the mathematical model, heuristic approach does not produce efficient results for the locations without direct link to the supply centers. The results support usage of the proposed mathematical model.