4. Revenue Management Flashcards
Fig. The Role of Revenue Management in Supply Chain Management
Definition of Revenue Managements
- encompasses a range of quantitative methods to support decision-making on accepting or rejecting uncertain, dynamic demand of varying value
- objective: use the inflexible and perishable capacity as efficiently as possible.
Price-Based Revenue Management
Pricing as primary decision variable
-> Dynamic pricing
Quantity-Based Revenue Management
Allocation of inventory and capacity as primary decision variable
-> Revenue Management in the strict sense
Instruments of Revenue Management
- Overbooking
- Differential Pricing
- Capacity allocation
Overbooking
▪ Compromise wasted capacity and capacity shortages
▪ Goal: Complete utilization of capacity despite uncertain demand
Differential Pricing
▪ Adjusting prices to meet the customer’s willingness to pay
▪ Goal: Exploit market potentials by forming segments with different willingness to pay
Capacity allocation
▪ Allocating capacities to different customer segments
▪ Goal: Maximizing profit by accepting or rejecting requests in anticipation of higher-price buyers arriving at a later point in time
Forming Customer Segments
- Temporal differentiation
– Time of purchase: Business client / tourist
– Time of use: Weekdays / weekends in hotels, cinemas, etc. - Channel / Regional differentiation
– e.g., price of a water bottle in the supermarket and at the airport - Flexibility
– Ability to cancel tickets, worldwide repair service for electronics - Group affiliation
– e.g., students, retirees, ADAC members, etc. - Product and service variations
– e.g., personal and commercial license for software, discount beyond certain trip length
Important: Differentiation needs to effectively prevent shifts between segments
Differential Pricing With Deterministic Demand
▪ Customers with a different willingness to pay which can be separated into multiple segments
(e.g., tourists and business travelers competing for airline seats)
▪ Assumption: Demand is price dependent & Demand function is known
▪ Objective: What price to charge for each segment to maximize total revenue?
Differential Pricing With Deterministic Demand - Mathematical Mode
𝐷𝑖 Demand of segment 𝑖
𝑝𝑖 Price for segment 𝑖
𝑐 Variable costs
𝑘 Number of segments
𝑄 Maximum capacity
𝐴, 𝐵 Demand function parameters
Capacity Allocation Under Uncertainty
▪ Customers have different willingness to pay and customer behavior in competing for capacity (e.g., Business / private customers)
▪ Idea: Reserve capacity (quota) for uncertain, high-priced demand
▪ Trade-off: If too much capacity is reserved for high-priced segments, capacity “expires” (spoilage). If too little capacity is reserved, profitable requests need to be declined (spill).
▪ Objective: Determine quota, such that revenue / profit is maximized
Littlewood’s Two-Class Model (1972)
Calculating Expected Marginal Revenue (EMR)
Assumptions:
▪ Fixed capacity
▪ Two classes of products (demand segments) with prices 𝑝1 > 𝑝2
▪ Uncertain demand (𝐷𝑖~𝑁(𝜇𝑖, 𝜎𝑖) 𝑓𝑜𝑟 𝑖 = 1,2)
▪ Demand for segment 2 is realized before demand for segment 1 realizes
→ How much capacity should be reserved for segment 1?
→ Optimal protection limit 𝑦1∗ is reached when the expected marginal profit of segment 1 is equal to the marginal profit of segment 2 (Littlewood’s Rule)
Littlewood’s Two-Class Model Solution
Multiple Customer Segments - Conclusion
▪ Companies with fixed capacities which serve multiple demand segments can increase their profitability by tactical pricing.
▪ Important: Fencing demand!
▪ Sufficient capacities should be reserved for high-priced demand (such that the expected marginal revenue corresponds to the marginal revenue of low-priced demand).
Dynamic Pricing (deterministic case)
▪ Perishable good, i.e., the value decreases over time (e.g., fruits, fashion articles, electronics, capacities)
▪ Assumptions: Demand is price-dependent and Demand function is known
▪ Idea: Exploit varying elasticities in demand over time to maximize revenue / profits
Dynamic Pricing (deterministic case) - Mathematical Model
𝑡 Period
𝑇 Number of periods
𝐷𝑡 Demand in period 𝑡
𝑝𝑡 Price for period 𝑡
𝑄 Maximum capacity
𝐴, 𝐵 Parameters of the demand function
Overbooking Under Uncertainty
▪ Customers often cancel their bookings, book at short notice only, or do not use a ticket they purchased
▪ Example Lufthansa: ca. 3 million “no-shows“ per year*
▪ This results into spoilage of the limited capacity and utilization decreases.
▪ Capacity released by “no-shows” is not known with certainty until shortly before the service is performed and can therefore no longer be sold
→ Idea: Overbooking the available capacity
▪ Objective: Minimizing the costs of wasted capacity in case of too many cancellations and costs of arranging a backup if too few cancellations lead to committed orders being higher than the available capacity
Fig. Overbooking Under Uncertainty - Optimal overbooking
𝑂∗ Optimal overbooking [units]
𝑠∗ Probability, that cancelations are smaller or equal to 𝑂∗
→ Customers need to be “put off”
𝐶𝑈 Costs of wasted capacity (insufficient number of reservations)
𝐶𝑂 Costs of capacity shortage (arranging a backup, rebooking/compensation)
𝜇𝑐, 𝜎𝑐 Distribution parameters of (normally distributed) cancellations
Perishable Goods Conclusion
▪ If demand sensitivity changes over time, dynamic pricing can be an effective instrument to increase profitability (e.g., articles of clothing)
▪ Overbooking or overselling of a supply chain asset is valuable if order cancellations occur and the asset is perishable
▪ The level of overbooking is based on the trade-off between the cost of wasting the asset if too many cancellations lead to unused assets and the cost of arranging a backup if too few cancellations lead to committed orders being larger than the available capacity
Capacity Allocation With Multiple Resources
Consider 𝑛 resources (bundle) with interdependencies between resources where one resource could limit the sales of a bundle.
How to select orders in a way that maximizes profitability?
→ Network Revenue Management
Bid-Price Controls
▪ Idea:
Calculating opportunity costs (as an acceptance threshold) by
multiplying the capacity utilization and dual prices (bid-prices) from the production planning program
▪ Accept a request only if:
▪ There are sufficient available capacities
▪ Revenue of request is greater or equal to opportunity costs
▪ Bid prices vary for each resource, at each point in time, and for each capacity level
Production Program Planning (single period) - Mathematical Model
𝑗 Products
𝑖 Resources
𝑎𝑖𝑗 Capacity utilization of resource 𝑖 for product 𝑗
𝑝𝑗 Price of product 𝑗
𝑐𝑖 Capacity of resource 𝑖
𝐷𝑗 Demand forecast for product 𝑗
𝑥𝑗 Production quantity of product j
Implementation aspects of Revenue Management
▪ Evaluate your market carefully
▪ Quantify the benefits of revenue management
▪ Implement a forecasting process
▪ Make revenue management decisions using optimization
▪ Involve both sales and operations
▪ Understand and inform the customer
▪ Integrate supply planning with revenue management
Aspects of Implementation - Intercompany / Supply Chain Perspective
▪ Managing demand leads to more stable resource requirements for suppliers
▪ Reduced effort of coordination because of lower demand variability
▪ Mitigation of bullwhip effect
▪ Increased utilization of supply chain assets
Aspects of Implementation - Company Perspective
▪ Decision support for dealing with the heterogenous requirements of different supply chains in B2B (e.g., automotive, aviation, and construction industry as customers of the steel industry)
▪ Increasing profitability
▪ Increasing delivery quality (individualized offer with respect to customer / product / time)
