Deep Study on Marketing Decision under the Price Fluctuation of Pig Market based on bayesian method

Bayesian method is a method that synthesizes the prior information about unknown parameters with sample information, then obtains the posterior information according to Bayesian formula, and then deduces the unknown parameters according to the posterior information. Markov decision process is the optimal decision process of stochastic dynamic system based on Markov process theory. In this paper, we establish a two-stage hierarchical Markov decision process, which simulates the sequential marketing decision under the fluctuation of pigsty price. The decision-making process is based on market information such as pork, piglets and feed prices. In addition, the bayesian method is used to update the information and embed it into the hierarchical Markov decision process, and the optimal strategy under different price fluctuation patterns is analyzed. We also evaluated the value of incorporating price information into the model. In the manufacture of fattening pigs, pig sales refer to a series of selection decisions until the production unit is emptied. The profit of a production unit depends on the price of pork, the cost of raising and the cost of buying piglets. The price fluctuation in the market will affect the profit, and the optimal marketing decision may also change under different price conditions. Most studies have considered pig marketing at constant prices. However, because price fluctuation can not have no effect on marketing decision, it is necessary to consider pig marketing under price fluctuation. In this paper, we establish a two-level hierarchical markov decision-making process that simulates sequential marketing decisions under price fluctuations in pigsty. The decision-making process is based on market information such as pork, piglets and feed prices. In addition, the bayesian method is used to update the information and embed it into the hierarchical markov decision process to analyze the optimal strategy in different price fluctuation modes. we also evaluated the value of incorporating price information into the model.