Someone familiar with prediction problems, where you need to predict how much goods to order given some history?
For example, this week you have the history of how much apples you ordered, how much potatoes, how much oranges, etc. and you need to predict how much to order for the next week?
This is an inventory management question, which has been very thoroughly studied in the field of operations research and supply chain management, and is called Inventory Theory. Since you have stated your problem simply, I will refer you to the most basic models in this area. If you like these, then you should know enough terminology to search effectively for more sophisticated models.
Economic Order Quantity Model: Simple model that forms the basis of the next few models. It is useful if you have relatively predictible/low uncertainty in the future demand for a product.
Stochastic Continuous-Reivew Models: Good for when there is considerable uncertianty in the demand. See Secs. 19.5 and 19.6 of this link to a very popular operations research textbook. You will find 19.6 particularly relevant to your problem, as it deals with perishable items.