In recent years, the traditional fresh fruit and vegetable (FFV) supply chain has been scrutinized due to, among other things, its high food waste generation and lack of flexibility to respond to market opportunities. One of the contributing factors to these problems is the lack of market information such as product demand and price data. The availability of this information will not only inform the supply chain participants of the current market, but it will also enable analyses on the future market to avoid overproduction, losses or small profits, product scarcity, and ultimately food waste. In particular, historical market prices for FFV are available, for more than 40 commodities and thirteen terminal market cities including Chicago, Los Angeles, and New York through the United States Department of Agriculture (USDA). Currently, the available USDA data is unstandardized, incomplete, and difficult for the supply chain participants to use. In addition, assessing the future market requires training a large quantity of models (commodities and markets) for which, not every analytical forecasting technique is adequate. Both constraints make it difficult for growers or other involved parties to use what limited price data is available. This work addresses this gap by developing a framework to automate the price prediction process. A generalized additive model is adapted for FFVs. Price variability caused by crop perishability, seasonality, and market differences were considered by conducting parameter tuning for each combination of crop and market. Initial validation was conducted from several years of USDA data. Furthermore, a desirability index is incorporated to select among competing models.