Fuzzy rule-based demand forecasting for dynamic pricing of a maritime company

In this study, the pricing problem of a transportation service provider company is considered. Our goal is to find optimal prices by using probabilistic dynamic programming. A fuzzy IF-THEN-rule based system is used to identify the demand levels under different prices and other characteristics of the journey. The results obtained by optimal price policies show that the revenue increases by applying dynamic pricing policy instead of fixed pricing. Thus, the diversification of pricing policies under different conditions is beneficial for the company. Dynamic pricing, often referred to as individual-level price discrimination , has become much more common with the increased usage of internet sales [21]. Internet provides a media for companies to offer different pricing strategies in which prices vary over time, consumers, and/or circumstances. Companies try to maximize their profits using these strategies. This attempt is often called as revenue management. The bulk of the literature deals with the revenue management and dynamic pricing problems of especially airlines, freight, and tourism markets. To the best of our knowledge, there has been no study conducted for maritime public transportation industry. In the transportation market, the demand fluctuates based on weather conditions, price, time of day and day of week. Based on these conditions, passengers decide on whether to travel or not. In other words, passengers' demand/price elasticity may change depending on the weekdays or weekend, early or late hours of a day as well as the weather conditions. Considering this fact, the strength of fuzzy logic in capturing the ambiguity and exploiting imprecise data created a motivation for us to use fuzzy logic in demand forecasting. Fuzzy IF-THEN rules are generated in order to retrieve the price/demand elasticity of passengers. Using these findings, a dynamic pricing model is formulated in order to find the optimal policies under different conditions. Then the policies obtained by dynamic pricing are compared with the fixed pricing policy. The paper includes seven sections. Literature review follows this section. In Section 3, problem description is given. Then dynamic pricing model is described in Section 4 while Section 5 presents the fuzzy rule based demand forecasting system. In Section 6, the case study is given and finally, conclusion and future research are discussed in Section 7. 2. Literature review Revenue management literature related to the transportation industry is reviewed below in order to reveal the contribution of this study. For a more detailed review of pricing models …