Managing Capacity and Shin Ming Guo Demand NKFUST Managing - - PDF document

1

Managing Capacity and Demand

• Managing dynamic demand
• Service capacity is perishable
• Yield Management

Shin‐Ming Guo NKFUST

Case: Disneyland Paris

 Established in 1992  Overestimate the demand  Too many hotel rooms lead to

high operating cost

 Need more space for bus parking  Unbalanced workforce scheduling

2

Service Capacity

 Participation: Need to be near customers  Simultaneity: Inability to transport services  Perishability: Inability to store services  Heterogeneity: Volatility of demand

Capacity: amount of output over a period of time Often use resource input to measure capacity

Focus: Matching Capacity with Demand

• Demand can vary and is unpredictable.
• Capacity is inflexible and maybe costly.
• Demand < Capacity  Impossible to stock service
• Demand > Capacity  Customers may not wait for service

4

3

Economic Consequences of Mismatch

5

Air travel Emergency Room Retailing Supply

Seats on specific flight Medical service Consumer electronics

Demand

Travel for specific time & destination Urgent need for medical service Kids buying video games

Supply Exceeds Demand

Empty seat Doctors, nurses, and infrastructure are under‐utilized High inventory costs

Demand Exceeds Supply

Overbooking; Profit loss Crowding and delays in the ER, Deaths Foregone profit; Consumer dissatisfaction

Capacity Utilization vs. Service Quality

Optimal operating level  70% of Design capacity

4

Matching Supply and Demand for Services

7

DEMAND Strategies 2 Partitioning demand 5 Developing complementary services 3 Differential Pricing 6 Developing reservation systems

12 Yield Management

Capacity Strategies 9 Cross‐ training employees 8 Franchising 7 Increasing customer participation 11 Scheduling work shifts 10 Using part‐time employees 1 Managing Variability 4 Promoting Off Peak Demand

• 1. Managing Customer-induced Variability

Type of Variability Accommodation Reduction Arrival Provide generous staffing Require reservations Capability Adapt to customer skill levels Target customers based on capability Request Cross‐train employees Limit service breadth Effort Do work for customers Reward increased effort Subjective Preference Diagnose expectations and adapt Persuade customers to adjust expectations

8

5

• 2. Segmenting Demand

20 40 60 80 100 120 140

• Mon. Tue. Wed. Thur.

Fri. Before Smoothing After Smoothing 9

Smoothing Demand by Appointment Scheduling Day Appointments Monday 84 Tuesday 89 Wednesday 124 Thursday 129 Friday 114

Too many walk‐in patients on Mondays at a health clinic.

• 3. Differential Pricing

10

6

• 4. Promoting Off-Peak Demand

 Different sources of demand

Hotel: conventions for business

• r professional groups during the
• ff‐season.

 Avoid waiting times

Department store: shop early and avoid the rush.

• 5. Developing Complementary Services
• A new service is the complementor if customers value your

service more when they already have purchased the existing service.

• Movie theaters offer popcorns and soft drinks.
• A new service is the complementor if it results in a more

uniform demand.

• Restaurants offer the “afternoon tea” service.
• Travel agency: Australia and New Zealand Tours

12

7

• 6. Reservation and Overbooking
• Taking reservations is like preselling the service.
• Reservations may benefit consumers by reducing waiting and

guarantee service availability.

• Approximately 50% of reservations get cancelled.
• Multiple reservations, late arrivals, no‐shows.
• Customers can cancel or postpone reservations— with a penalty
• Airlines and hotels can overbook reservations— with a penalty

13

Overbooking to Protect Revenue

Overbooking—accept more reservations than supply Example: On average there would be 10 cancellations or no‐

• shows. So the hotel can accept 10 more reservations.

14

Too much overbooking: some customers may have to be denied a seat even though they have a confirmed reservation. Too little overbooking: waste of capacity, loss of revenue

8

Example: Surfside Hotel

15

expected number of no‐shows = 0(0.07)+1(0.19)+…+9(0.01)=3.04 Expected opportunity loss = 3.04 × $40 = $121.60

16

Cost of too many overbooking: Co=$100 for accommodation at some other hotel and additional compensation. Cost of not enough overbooking: Cu=$40 per room.

9

Overbooking Solution

• Critical ratio
• Find x such that x is the largest number that satisfies

P(number of no‐shows < x) ≤ 0.286

• Optimal number of overbooking = 2
• There is about a 26% chance that the hotel will have more

customers than rooms.

17

286 . 100 40 40    

• u

u

C C C

Strategies for Managing Capacity

18

10

• 7. Customer Participation

Customer participates actively in the service process.

Objectives:

• Cost reduction (less personnel is needed)
• Capacity becomes more “variable”, according to demand

Disadvantages:

• Customer expects quicker service
• Customer expects low prices (compensation for his help)
• Quality of customers “work” cannot be controlled by company

(e.g., customer can leave his waste on the table)

19

• 8. Franchising

Benefits to the Franchisor Less financial investment Quick expansion to other markets Economies of Scale Problems Franchisee does not receive proper training Franchisee fails to follow the contract or regulations Franchisor does not have new product development Franchisor fails to provide support

11

Economies of Scale for Service Industry

• Chain stores lead to buying power.
• Travel agency buy airline tickets and hotel

rooms in bulks to get deeper discount.

• Small business can form an alliance to

increase the bargaining power against big suppliers.

 Competing retail stores or restaurants located in the

same area may attract more consumers.

 Commuter cleaning  Economies of scale may hurt service quality

• 9. Cross-training & Part-time Employees

Training employees to be able to do different tasks

• Demand peaks: Each employee performs his specialized work

(e.g., cashier in a supermarket)

• Low demand: Employee performs additional tasks: Job is

enlarged (e.g., filling the shelves in a supermarket)

Using part‐time employees

• When demand peaks can be foreseen: Additional staff can be

employed for these times (e.g., lunchtime in restaurants)

• Skills needed low: Students can be taken (e.g., bakery)

12

• 10. Adjustable Capacity
• Airlines: Different aircrafts
• Rental Cars: ability to move cars around.

Workshift Scheduling

• The peak to valley variation is 125 to 1.
• Carefully schedule the workforce so that

the required service level can be maintained with the minimal cost.

23

Convert Demand and Schedule Shifts

24

13

Scheduling Consecutive Days Off

25

Mon Tue Wed Thu Fri Sat Sun forecast 4 3 2 4 3 1 2 4 3 2 4 3 1 2 3 2 1 3 2 1 2 2 1 2 2 1 1 1 1 1 1

A B C D

• 12. Revenue Management
• Return = Revenue – Operations Cost

= Throughput  Price – Fixed Costs –Throughput  Variable Costs – Reduce fixed costs – Reduce variable costs – Increase price – Increase throughput

• If capacity is fixed and perishable, fixed costs are high and

variable costs are low, increasing price and/or throughput to improve profitability.

26

14

Case: Increase Revenue with Fixed Capacity

• The Park Hyatt Philadelphia, 118 King/Queen rooms.
• Regular fare is rH= $225 (high fare) targeting business travelers.
• Hyatt offers a rL= $159 (low fare) discount fare for a mid‐week

stay targeting leisure travelers.

• Demand for low fare rooms is abundant.
• Most of the high fare demand occurs
• nly within a few days of the actual stay.

27

Booking Limits and Yield Management

• Choice 1: Do not accept low fare reservation.

Hope that high fare customers will eventually show up.

• Choice 2: Accept low fare reservations

without any limit.

• Choice 3: Accept low fare reservations but

reserve rooms for high fare customers

• Objective: Maximize expected revenues by

controlling the sale of low fare rooms.

28

15

Yield Management: Airline Pricing

29

• Carriers typically fill 72.4% of seats and have a break‐even

load of 70.4%.

• Very high fixed costs and perishable capacity.

Example: Blackjack Airline

30

d = demand for full fare ($69) ~ N(60, 152) Expected revenue=6960=$4140 Demand for “gamblers fare” ($49) is abundant Expected revenue=4995=$4655 Decision: x = seats reserved for full fare passengers

95 seats

16

Optimal Booking Solution

• (z)=P(d < x)=0.29  z= -0.55

31

) 1 , ( ~ 15 60 N d d z       51 15 ) 55 . ( 60 55 . 15 60          x x z 29 . 49 20 20 ) (      

• u

u

C C C x d P

Cost of too many full fare seats reserved: Co=$49 Cost of not enough full fare seats reserved: Cu=$20

Optimal Revenue for Blackjack Airline

• Z= ‐0.55  Normal Loss Function L(z)

=NORMDIST(z,0,1,0)‐z*(1‐NORMSDIST(z)) =0.7328

• For full fare customer

expected loss (due to not enough seats reserved) =L(z)∙=0.7328=10.99 expected sales + expected loss = expected full fare demand  expected sales=expected demand‐expected loss =60‐10.99=49.01

• Expected total revenue=49.01*69+(95‐51)*49 =$5537

17

Yield Management for a Resort Hotel

33

Ideal Characteristics for Yield Management

• Relatively Fixed Capacity
• Ability to Segment Markets
• Perishable Inventory
• Product Sold in Advance
• Fluctuating Demand
• Low Marginal Sales Cost and High Capacity Change Cost

34