Sie sind auf Seite 1von 68

Chapter 11

Managing Capacity and Demand


McGraw-Hill/Irwin
Service Management: Operations, Strategy, and
Information Technology, 6e
Copyright 2008 by The McGraw-Hill Companies, Inc. All Rights Reserved.
Learning Objectives
Describe the strategies for matching capacity
and demand for services.
Recommend an overbooking strategy.
Use Linear Programming to prepare a weekly
workshift schedule.
Prepare a work schedule for part-time
employees.
Explain what yield management is and how it
is applied.
11-2
Introduction
After fixed capacity investment
decisions have been made (e.g.,
number of hotel rooms to be built or
aircraft to be purchased), the hotel
beds must be filled or the airline seats
sold to make the daily operations
profitable.

Introduction-2
The challenge is faced by managers of
matching service capacity with
customer demand on a daily bases in a
dynamic environment.
Service capacity is a perishable
commodity
A service is produced and consumed
simultaneously.
Introduction-3
Whenever demand for a service fall
short of the capacity to serve, the
results are idle servers and facilities.
The variability in service demand is
quite pronounced, and, in fact our
culture and habits contributes to these
fluctuations.
Generic Strategies of Level
Capacity or Chase Demand
There are two generic strategies for
capacity management:
1. Level capacity
2. Chase capacity
Table 11.1 (next slide) illustrates the
trade-offs between these two strategies


Generic Strategies of Level
Capacity or Chase Demand-2
Level Capacity: we introduce marketing
oriented strategies such as price incentives
that can smooth customer to better utilize
the fixed capacity (e.g., utilities)
Chase strategy: we consider operation -
oriented strategies such as workshift
scheduling to vary capacity to mach the
changing levels of customer demand (e.g.,
call centers)
Generic Strategies of Level
Capacity or Chase Demand-3

Hybrid Strategy: e.g., facility capacity in hotel
beds is fixed but staffing can vary according
to seasonal demand.
Figure 11.1 shows the strategies for demand
management when a level capacity is being
maintained and strategies for capacity
management when a chase demand is being
pursued.
Level Capacity and Chase Demand
(Table 11.1)
Strategic Dimension Level Capacity Chase Demand
Customer Waiting Generally Low Moderate
Employee Utilization Moderate High
Labor-skill Level High Low
Labor Turnover Low High
Training Required per Employee High Low
Working Conditions Pleasant Hectic
Supervision Required Low High
Forecasting Long-run Short-run
11-9
Strategies for Matching Capacity
and Demand for Services

MANAGING
DEMAND
Partitioning
demand
Developing
complementary
services
Establishing
price
incentives
Developing
reservation
systems
Promoting
off-peak
demand
Yield
management
MANAGING
CAPACITY
Cross-
training
employees
Increasing
customer
participation
Sharing
capacity
Scheduling
work shifts
Creating
adjustable
capacity
Using
part-time
employees
11-10

Customer-induced Variability
Arrival: customer arrivals are independent
decisions not evenly spaced.
Capability: level of knowledge and skills vary
resulting in some hand-holding.
Request: uneven service times result from
unique demands.
Effort: level of commitment to coproduction
or self-service varies.
Subjective Preference: personal preferences
introduce unpredictability.
11-12
Segmenting Demand
Demand for a service seldom drives
from a homogeneous source
Demand often is grouped into random
arrivals and planned arrivals.
Segmenting Demand:Health
Clinic Demand
An Analysis of health clinic demand showed
that the greatest number of walk-in patients
arrived on Monday and fewer during the
remaining weekdays.
While walk-in demand is not controllable, app
ointments are
Therefore, why not make appointments in the
latter part of the week to level demand ?
Segmenting Demand:Health
Clinic Demand-2
Using data for the same week in the previous
year, these researchers noted the number of
walk-in patients for each weekday.
Subtracting these walk-in patients from daily
physician capacity gives the number of
appointment patients who are needed each
day to smooth demand
Fig.11.2 Effect of Smoothing
Physician Visits

Segmenting Demand at a Health Clinic
0
20
40
60
80
100
120
140
Mon. Tue. Wed. Thur. Fri.
Before
Smoothing
After
Smoothing
Smoothing Demand by Appointment
Scheduling

Day Appointments

Monday 84
Tuesday 89
Wednesday 124
Thursday 129
Friday 114
11-17
Offering Price Incentives
Examples of different pricing
1. Weekend night rates for long distance
telephone calls
2. Matinee or reduced prices before 6 pm at
movie theaters
3. Off-season hotel rates at resort locations
4. Peak-load pricing by utility companies
Offering Price Incentives-2
Differential pricing has been suggested
for federal campsites to encourage
better use of this resource.
A company developed a discriminatory
pricing system to ensure that camping
fees accurately reflect the marginal
benefit of the last campsite on any
given day
Offering Price Incentives-3
They identified four different camping
experiences on the basis of days and
weeks of the camping season.
Table 11.3 contains a schedule of daily
fees by experience type.

Discriminatory Pricing for Camping
Experience No. of Daily
Type Days and weeks of camping season Days Fee
1 Saturdays and Sundays of weeks 10 to 15, plus 14 $6.00
Dominion Day and civic holidays
2 Saturdays and Sundays of weeks 3 to 9 and 15 to 19, 23 2.50
plus Victoria Day
3 Fridays of weeks 3 to 15, plus all other days of weeks 43 0.50
9 to 15 that are not in experience type 1 or 2
4 Rest of camping season 78 free

EXISTING REVENUE VS PROJECTED REVENUE FROM DISCRIMINATORY PRICING

Existing flat fee of $2.50 Discriminatory fee
Experience Campsites Campsites
Type occupied Revenue occupied (est.) Revenue
1 5.891 $14,727 5,000 $30,000
2 8,978 22,445 8,500 21,250
3 6,129 15,322 15,500 7.750
4 4,979 12,447 . .
Total 25,977 $ 64,941 29,000 $59,000
11-21
Offering Price Incentives-4
During the 78 days of free camping,
however, a saving in labor cost is
possible, because no ranger is needed
to collect fee at the campsite.
Even so, for the arrangement to work
effectively in altering demand, it must
be well advertised and include an
advance booking system for campsite.
Offering Price Incentives-5
Note the projected increase in demand for
experience type 3 because of the
substantially reduced fee.
The result of-peak pricing is to tap a latent
demand for campsites instead of
redistributing peak demand to off-peak times.
Thus, discriminatory pricing fills in the valleys
instead of leveling off the peaks.
Promoting Off-Peak Demand
Creative use of off-peak capacity results from
seeking different sources of demand.
One example is use of a resort hotel during
the off-season as a retreat location for
business or a professional groups.
The strategy of promoting off-peak demand
can be used to discourage overtaxing the
facility at other times.
Developing Complementary
Services
Complementary services being offered to
occupy witing customers.
Restaurants have discovered the benefits of
complementary services by adding a bar.
Developing complementary is a natural way
to expand ones market, and it is particularly
attractive if the new demands for service are
contracyclical and result in a more uniform
aggregate demand. E.g., heating contractors
also perfform air-conditioning units.
Reservation Systems and
Overbooking
Taking reservations presales the potential
service.
As reservations are made, additional demand
is deflected to other time slots at the same
facility or to other facilities within the same
organizations.
Reservations also benefit consumers by
reducing waiting and guaranteeing service
available.
Resevation Systems and
Overbookin-2
Problem do arise, however, when
customers fail to honor their
reservations. (these customers are
referred as no-shows)
Faced with flying empty seats because
of the no-shows, airlines adopted a
strategy of overbooking.
Reservation Systems and
Overbooking-3
A good overbooking strategy should minimize
expected opportunity cost of idle service
capacity as well as the expected cost of
turning away reservation.
Adopting this strategy requires training
frontline personnel to handle graciously
guests whose reservations can not be
honored
Reservation Systems and
Overbooking: Example 11.1
During the past tourist season, Surfside Hotel
did not achieve very high occupancy despite
a reservation system that was designed to
keep the hotel fully booked. Apparently,
prospective guests were making reservations
that, for one reason or another, they failed to
honor. A review of front-desk records during
the current peak period, when the hotel was
fully booked, revealed the record of no-show
given in Table 11.5.
Reservation Systems and
Overbooking: Example 11.1
Table 11.6 displays the loss that is associated
with each possible overbooking alternative
For each overbooking strategy, the expected
loss is calculated by multiplying the loss for
each no-show possibility by its probability of
occurrence and than adding the products.
Hotel Overbooking Loss Table
Number of Reservations Overbooked
No- Prob-
shows ability 0 1 2 3 4 5 6 7 8 9
0 .07 0 100 200 300 400 500 600 700 800 900
1 .19 40 0 100 200 300 400 500 600 700 800
2 .22 80 40 0 100 200 300 400 500 600 700
3 .16 120 80 40 0 100 200 300 400 500 600
4 .12 160 120 80 40 0 100 200 300 400 500
5 .10 200 160 120 80 40 0 100 200 300 400
6 .07 240 200 160 120 80 40 0 100 200 300
7 .04 280 240 200 160 120 80 40 0 100 200
8 .02 320 280 240 200 160 120 80 40 0 100
9 .01 360 320 280 240 200 160 120 80 40 0
Expected loss, $ 121.60 91.40 87.80 115.00 164.60 231.00 311.40 401.60 497.40 560.00

11-31
Overbooking Management Using
the Critical Fractile Model

P(d<x) C
u
/(C
u
-C
o)
Where,
C
u
= the number of no-show under
estimated ($40 room contribution
that is lost when a reservation is not
honored
C
o
= the number of no-shows is overestimated (the
$100 opportunity loss associated with not having a
room available for an overbooked guest)
d = the number of no-shows based on past experience
x = the number of rooms overbooked.
Overbooking Management Using
the Critical Fractile Model-2
Thus the number of rooms overbooked
should just cover the cumulative probability
of no shows and no more, as calculated
below:
P(d<x) $40/($40+$100) 0.286
From Table 11.5, a strategy of overbooking
by two rooms satisfies the fractional criterion,
because the cumulative probability
P(d<x)=0.26
Strategies for Managing
Capacity:Defining Service
Capacity
Service capacity is defined in terms of
an achievable level of output per unit
time (e.g., transactions per day for a
busy bank teller)
Service capacity also can be defined in
terms of the supporting facility, such as
number of hotel bed or available seat
miles at the system level for airline.
Strategies for Managing
Capacity:Defining Service
Capacity-2
For many services, demand cannot be smoothed very
effectively.
Consider, for example, demand at a call center as
shown in Fig. 11.3.
These data are the half-hourly call rates during a
typical 24-hour day.
The peak-to-valley variation is 125 to 1.
No inducement are likely to change this demand
pattern substantially
Control must come from adjusting service capacity.
Several strategies can be used to achieve this goal.

Strategies for Managing
Capacity: Daily Workshift
Scheduling
Workshift scheduling is an important
staffing problem for many
organizations the face cyclical
demand, such as telephone
companies, hospitals, banks and
police department.

Strategies for Managing
Capacity: Daily Workshift
Scheduling-2
The general approach is given below:
1. Forecast demand by hour
2. Convert the forecasted demand to hourly service
stuffing (operator) requirements
3. Schedule of tours, or Shifts is developed to match
the stuffing requirements profile as closely as
possible.
4. Assign operators to tours or shifts; specific service
personnel are assigned to tours, or shifts
Strategies for Managing
Capacity: Daily Workshift
Scheduling-3
Illustration of the approach on call center
problem:
1. Forecast Demand: Daily demand is forecast in
half-hour interval as shown in Fig. 11.3, and
must account for both weekday and weekend
variations as well as seasonal adjustments.
2. Convert to Operator Requirements: A profile of
half-hour operator requirements is developed on
the basis of the forecast daily demand and call
distribution.
Strategies for Managing
Capacity: Daily Workshift
Scheduling-4
The agreed service level requires that
89% of the time, an incoming call can be
answered within 10 seconds. The half-
hour operator requirements are
determined using a conventional queuing
model to ensure that the service level is
achieved for each half-hour. The result is
a top-line profile of operators required by
the half-hour, as shown in Figure 11.4.


Strategies for Managing
Capacity: Daily Workshift
Scheduling-5
3. Schedule Shifts: Tours, representing various
start and end times of work, need to be
assigned so that they aggregate to the top
line profile, shown in Fig. 11.4. A heuristic
computer program prepared especially for
this problem chooses tours from the
permissible set such that the absolute
difference between operator requirements
and the operator assigned is minimized when
summed over all n half-hour periods
Minimize | R
i
- W
i
|
Strategies for Managing
Capacity: Daily Workshift
Scheduling-6
The schedule building process is
shown schematically in Fig.11.5. At
each iteration, one tour at a time is
selected from all possible tours.
selected tour is the one that best
meets the criterion stated in the
above equation.
Figure 11.5 The Schedule
Building Process

Strategies for Managing
Capacity: Daily Workshift
Scheduling-7
4. Assign Operators to shifts: Given the set
of tours required, of the assignment of
operators to these tours is complicated,
because of the 24-hours, 7-days-per-
week operation.
Daily Scheduling of
Telephone Operator Workshifts

0
5
10
15
20
25
30
Time
N
u
m
b
e
r

o
f

o
p
e
r
a
t
o
r
s
Scheduler program assigns
tours so that the number of
operators present each half
hour adds up to the number
required

Topline profile
12 2 4 6 8 10 12 2 4 6 8 10 12
Tour
0
500
1000
1500
2000
2500
Time
C
a
l
l
s
12 2 4 6 8 10 12 2 4 6 8 10 12
11-44
Strategies for Managing Capacity:
Weekly Workshift Scheduling with
Days-Off Constraint
A typical employee works 5 days a week with
2 consecutive days off each week, but not
necessarily Saturday and Sunday.
Management is interested in developing work
schedules and meeting the varying employee
requirements for weekdays and weekends
with the smallest number of stuff members
possible.
The problem can be formulated as an integer
linear programing (ILP) model
Strategies for Managing Capacity:
Weekly Workshift Scheduling with
Days-Off Constraint-2

x
b
Example 11.2 Hospital
Emergency Room
The emergency room is operated on a
24 hour, 7 days-per-week schedule.
The day is divided into three 8-hour
shifts. The total number of nurses
required during the day shift is:

Day Su M Tu W Th F Sa
Nur
ses
3 6 5 6 5 5 5
Example 11.2 Hospital
Emergency Room-2
The emergency room director is
interested in developing a workforce
schedule that will minimize the number
of nurses required to staff the facility.
Nurses work 5 days a week and are
entitled to two consecutive days-off
each week.
Example 11.2 Hospital
Emergency Room-3
The ILP model above is formulated with the
appropriate RHS constraint values (i.e., b
1
=2,
b
2
=6.., b
7
=5).
The solution yields the following results: x
1
, x
2
=1,
x
3
=2, x
4
=0, x
5
=3, x
6
=0, x
7
=1.
The solution implies that we have one tour with
Sunday and Monday off, one tour with Monday and
Tuesday off, two tours with Tuesday and Wednesday
off, three tours with Thursday and Friday off, and
one tour with Saturday and Sunday off.
The optimal stuffing schedule is shown in table 11.7
with excess staff occurring only on Sunday and
Saturday.

These scheduling problems typically
result in multiple optimal solutions.
In this case, the solution x
1
=1, x
2
=1,
x
3
=1, x
4
=1, x
5
=1, x
6
=1, x
7
=2 is
feasible and also requires eight
nurses. Why might this second
solution be preferred to the schedule
shown in Table 11.7 ?




Example 11.2 Hospital
Emergency Room-4
LP Model for Weekly Workshift Schedule
with Two Days-off Constraint
(Emergency Room Problem, Table.11.7)

Objective function:
Minimize x
1
+ x
2
+ x
3
+ x
4
+ x
5
+ x
6
+ x
7
Constraints:
Sunday x
2
+ x
3
+ x
4
+ x
5
+ x
6
> 3
Monday x
3
+ x
4
+ x
5
+ x
6
+ x
7
> 6

Tuesday x
1
+ x
4
+ x
5
+ x
6
+ x
7
> 5

Wednesday x
1
+ x
2
+ x
5
+ x
6
+ x
7
> 6
Thursday x
1
+ x
2
+ x
3
+ x
6
+ x
7
> 5
Friday x
1
+ x
2
+ x
3
+ x
4
+ x
7
> 5
Saturday x
1
+ x
2
+ x
3
+ x
4
+ x
5
> 5

x
i >
0 and integer
Schedule matrix, x = day off

Nurse Su M Tu W Th F Sa
1 x x ...
2 x x
3 ... x x
4 ... x x
5 x x
6 x x
7 x x
8 x x
Total 6 6 5 6 5 5 7
Required 3 6 5 6 5 5 5
Excess 3 0 0 0 0 0 2

11-51
Strategies for Managing capacity:
Increasing Customer Participation
It is illustrated best by the fast-food
restaurants that have eliminated
personnel who serve food and clear
tables.
Capacity to serve varies more directly
with demand rather than being fixed.
Strategies for Managing capacity:
Creating Adjustable Capacity
Through design, a portion of capacity can be
made variable.
For example, airlines routinely move the
partition between first class and coach to
meet the changing mix of passengers.
Capacity at peak periods can be expanded by
the effective use of slack times.
Strategies for Managing
capacity: Sharing Capacity
A service delivery system often requires
a large investment in equipment and
facilities.
During periods of underutilization, it
may be possible to find other uses for
this capacity.
Strategies for Managing capacity:
Cross Training Employees
Some systems are made up of several
operations.
When one operation is busy, another
operation sometimes may be idle
Cross training employees to perform
tasks in several operations creates
flexible capacity to meat localized peaks
in demand.
Strategies for Managing capacity:
Using Part Time Employees
When peaks of activity are persistent
and predictable, part time help can
supplement regular employees.
Another source of part-time help is off-
duty personnel who are placed on
standby.
Strategies for Managing capacity:
Scheduling Part-Time Tellers at a
drive-In Bank
Derive-in banks experiences in activity on different
days of the week.
Fig.11.6 shows the teller requirements for a typical
week based on customer demand variations.
To reduce teller costs, management decided to
employ part time tellers, and to reduce the full time
staf to a level that just meets the demand for
Tuesday.
To provide equity in hours worked, it was decided
that a part time teller should work at least 2, but no
more than 3, days in a week
Strategies for Managing capacity:
Scheduling Part-Time Tellers at a
drive-In Bank-2
A primary objective of scheduling part
time workers is to meet the
requirements with a minimum number
of teller-days.
A secondary objective is to have a
minimum number of part-time tellers.
This approach is illustrated here using
bank tellers:
Strategies for Managing capacity:
Scheduling Part-Time Tellers at a
Drive-In Bank-3
1. Determine the Minimum Number of
Part Time Tellers Needed
2. Develop a Decreasing-Demand
Histogram
3. Assign Tellers to the Histogram
Scheduling Part-time Bank Tellers

Objective function:
Minimize x1+ x2+x3+x4+x5+x6+x7
Constraints:
Sunday x2+x3+x4+x5+x6 > b1
Monday x3+x4+x5+x6+x7 > b2











0




1




2




3




4






5




6




7

T
e
l
l
e
r
s

r
e
q
u
i
r
e
d

Mon. Tues. Wed. Thurs. Fri.
Two Full-time Tellers
5
4
1
3
2
1
4
3
2 1
5 2
Fri. Mon. Wed. Thurs Tues.




0



1



2




3




4



5

T
e
l
l
e
r
s

r
e
q
u
i
r
e
d

Decreasing part-time teller demand histogram
DAILY PART-TIME WORK SCHEDULE, X=workday

Teller Mon. Tues. Wed. Thurs. Fri.
1 x . x . x
2 x . . x x
3,4 x . . . x
5 . . x . x
11-60

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
11-62
Seasonal Allocation of Rooms by
Service Class for Resort Hotel

First class


Standard


Budget

P
e
r
c
e
n
t
a
g
e

o
f

c
a
p
a
c
i
t
y

a
l
l
o
c
a
t
e
d

t
o

d
i
f
f
e
r
e
n
t

s
e
r
v
i
c
e

c
l
a
s
s
e
s

60%
50%
30%
20%
50%
Peak Shoulder Off-peak Shoulder
(30%) (20%) (40%) (10%)
Summer Fall Winter Spring

Percentage of capacity allocated to different seasons
30%
20% 20%
10%
30%
50%
30%
11-63
Demand Control Chart for a Hotel

0
50
100
150
200
250
300
350
1 5 9 13 17 21 25 29 33 37 41 45 49 53 57 61 65 69 73 77 81 85 89
Days before arrival
R
e
s
e
r
v
a
t
i
o
n
s
Expected Reservation Accumulation
2 standard deviation control limits
11-64
Yield Management Using the
Critical Fractile Model


P d x
C
C C
F D
p F
u
u o
( )
( )
< s
+
s

Where x = seats reserved for full-fare passengers


d = demand for full-fare tickets
p = proportion of economizing (discount) passengers
C
u
= lost revenue associated with reserving one too few seats
at full fare (underestimating demand). The lost opportunity is the
difference between the fares (F-D) assuming a passenger, willing
to pay full-fare (F), purchased a seat at the discount (D) price.
C
o
= cost of reserving one to many seats for sale at full-fare
(overestimating demand). Assume the empty full-fare seat would
have been sold at the discount price. However, C
o
takes on two
values, depending on the buying behavior of the passenger who
would have purchased the seat if not reserved for full-fare.
if an economizing passenger
if a full fare passenger (marginal gain)

Expected value of C
o
= pD-(1-p)(F-D) = pF - (F-D)

C
D
F D
o
=

( )
11-65
Figure 11.11 Critical Fractile
for Blackjack Airlines

Topics for Discussion
What organizational problems can arise from the use of
part-time employees?
How can computer-based reservation systems increase
service capacity utilization?
What possible dangers are associated with developing
complementary services?
Will the widespread use of yield management
eventually erode the concept of fixed prices?
Go to http://en.wikipedia.org/wiki/Yield_management/
and discuss the ethical issues associated with yield
management.
11-67
Interactive Exercise
Watch the PowerPoint presentation
concerning the overbooking experience at
the Doubletree Hotel in Houston, Texas.
How could this situation been handled
differently?
11-68

Das könnte Ihnen auch gefallen