Conditions of Revenue Management

Business Condition
Transatlantic passenger ships are nearly killed off by the air travel. However, marine passenger transportation is still flourishing in the form of cruise ship (Marc, 1989, p.38). According to the Cruise Line International Association (CLIA, 2005), more than 100 million North Americans took cruise between 1980 and 2004. Besides, the average annual passenger growth rate of cruise line industry is 8.2 percent due the cruise industry report (Mazzarella& Ji, 2007, p.20), which indicates that cruise industry becomes one of the fastest growing sectors in the leisure travel market. Additionally, the cruise industry has grown as new markets that are developed and new ships are particularly designed for the specific cruise destination (Marc, 1989, p.38).
Precondition
The application of revenue management is not appropriated in all the industries. According to Kimes (1989), successful industry to apply revenue management must fit with RM characteristics, which in terms of perishable inventory, fixed capacity, market segmentation, advanced sales, low marginal costs and time-variable demand (cited in IDeaS, 2005, p.4). Kimes developed a typology model of revenue management as figure1, which includes two strategic levers of duration and price. Cruise industry would be suitable and be really successful to apply revenue management as both the capacity and duration are managed (Kimes, 2000, p.6). Kimes (1989, 1998) and Cross (1998) together complement five “necessary ingredients” for the successfully implementation of Yield management, which in terms of “ market segmentation, historical demand and booking pattern, pricing knowledge, overbooking policy and information systems.” However, Schwartz’s (1998) state that the key elements in yield management actually is only perishability of the product and customer’s willingness to pay (cited in Julian, 2000, p.295), which indicates Kimes and Cross are overstated. Conversely, Julian (2000) points out that

Cited: in Julian, 2000, p.295), which because a numbers of unique features make cruise even more complex and quite dissimilar from the Yield Management point of view (Julian, 2000, p.294). Similar to other transportation industry, once the property has departed, all unsold inventory became no value (IDeaS, 2005, p.5). Cruise ships’ capacity not only limited by number of cabins, but also the lifeboat seat capacity. This is so because the maximum of the guest can stay on the ship is based on the lifeboat seat rather than number of cabins, which restricted by law (Biehn, 2006, p.135). Comparatively, cruise line is dominated by leisure customers (Mazzarella & Ji, 2007, p.21), who are more price sensitive (IDeaS, 2005, p.10). On the other hand, leisure customers are less time sensitive, therefore, cruise line revenue manager enjoying long booking leading time that allowed them to operate in extended time frame (Mazzarella & Ji, 2007, p.21). Figure1 (cited in Kimes, 2000, p.6) RM models As aforementioned, on of the essential difference between cruise line industry and other tourism industry is “the capacity has two dimensions, which is number of cabin and lifeboat seats” (Li, 2010). Biehn(2006) formulates a deterministic liner program model considering the constraint of lifeboat capacity. However, Li (2010) points out the model is too simple and limited in each product only contain one cabin and at least guest. Li (2010) and Mazzarella&JI (2007) state static model and dynamic model are two fundamental approaches to solve the complex situation which under the environment of demand uncertainty and the constraints of two-dimensional capacity. Static model is widely used because it is simple and less data required. However, static model limited in optimal from a revenue maximization perspective (Mazzarella&Ji, 2007,p.23). Mazzarella&Ji (2007) modified Belobaba (1989)’s EMSR model into a new model- NCA (nested class association), and developed DCA model (dynamic class association) based on the system of Lee and Hersh (1993) and the algorithm of Shy (2005). Mazzarella&Ji demonstrate that FCFS method only generate more revenue when the demand is lower than the booking limits. Moreover, Mazzarella&JI believed by applying both NCA and DCA with good demand forecast can anticipate an average 4.2 to 6.3 per cent increase in revenue which should fall to bottom line profitability. LI (2010) construct static model with four solve method which in terms of chance constrained programming, robust optimization, deterministic programming, and bid-price control, and a dynamic capacity allocation model by applying Markov Decision Process to maximize the total expected revenue from all kinds of cruise products to solve the cruise two-dimensional revenue management problem under uncertain environment. Additionally, Li (2011) develops risk decision model of cruise line overbooking system that can be solved by a real option approach. He comments that revenue management could be regard as a part of risk decision in the business operation. In the cruise line industry, real options approach is a strong risk decisions tool deal with the overbooking or denied boarding risk control (Li, 2011, p.157). Implementation of RM in Cruise Line Industry Kimes(1989,1998) and Cross (1997,1998) has defined capacity management as an essential element to implement yield management efficient (cited in Julian, 2000, p.297), which is the most complex in the operation of cruise line industry than any other tourism industry (Julian, 2000, P.298). The main objective of revenue management process in cruise line industry is to maximize profit within in the fixed the capacity (cabin and lifeboat seat). The revenue management model that discussed above can help the cruise manager make a right decision on receiving or rejecting the reservation either for singles or families to manage the cruise’s capacity effectively. However, Julian (2000), LI (2010, 2011) and Mazzarella&JI(2007) point out that model only works effectively based on the sufficient historic data and accuracy forecasting. Booking curve shows the increase in reservation overtime, which is the foundation of any yield management system (Kimes, 2000,p.10). Biehn(2006) mentioned that January to March is the wave demand period in cruise industry, which active 30-40% reservation all through the year. However, the wave booking period is not the most popular period to take cruise but the time of year bookings occur and not only the sail date the passenger plans to actually board the ship, which should be differentiate from almost all industries that use revenue management, seasonality patterns occur when the resources are used (Biehn, 2006, p.139). By knowing the booking pattern allows manager be better able to decide on reservations to accept or to deny (Kimes, 2000, p.10). Developing logical differential pricing polices is also essential to use revenue management effectively. Kimes (2000) points out “logical” is important as the customer should be able to see the distinction between different prices being quoted (e.g. in hotel industry, different room types & different view), otherwise, the differential price strategy may not work. Rate fence wildly used to justify price discrimination. However, cruise line industry typically has different cabin categories all at different prices. Therefore, customers are always “buy up” or “buy down” for the types of accommodation since the features and price are similar (Biehn, 2006, p.136). This is so may because, cruise line industry not pricing on the cabin but price and manage each single customer separately (Biehn, 2006, p.135). A logical overbooking tool is a necessary tool to apply revenue management effectively. Cruise line companies could overbook to protect themselves against the possibility of no-shows (Kimes, 2000, p.10), however, in the cruise industry typically, the implementation of overbooking causes denied boarding, economic losses which directly related to the denied boarding, and even the company brand image will be damaged (Li, 2011, p.156). In this regards, cruise line companies must collect information on historical no show and cancellation rate, and on the other hand, train the employee well on how to deal with the displaced customer (Kimes, 2000, p.10).