MEI has approximately ten Stores spread across South Sumatra, which are supplied by three Distribution Centers, namely DC Palembang, DC Jambi, and DC Bengkulu. In its distribution, the constraints are the distance between several shipments to each store, which takes a long time and costs a lot of money. Based on these conditions, the purpose of this study is to balance the supply volume and determine the optimal route. For this study, forecasting will be carried out using the ARIMA and Croston methods because of the sporadic sales level. Furthermore, optimization of the DC supply location points and supply volume is carried out using linear programming, and the best route is determined using mixed integer linear programming with the Northwest Corner method. The forecasting results are used to predict the total supply load at three depots in the South Sumatra region; then, the supply volume is determined using linear programming. It is known that the Store locations that DC Palembang will supply are Muara Enim, Pagar Alam, Banyuasin, and Lahat. DC Jambi will supply Bungo, Batang Hari, and Kerinci Stores, while DC Bengkulu, which previously only supplied 2 Stores, now has 4 Stores, including Pagar Alam, Lubuk Linggau, Lebong, and Kepahiang. Based on the regional transportation model of South Sumatra, transportation costs can be minimized by up to 21%.