Capacity Planning and

Queuing
UNIT 5
 Capacity Planning Challenges
• Inability to create a steady flow of demand to fully
utilize capacity
• Idle capacity always a reality for services.
• Customer arrivals fluctuate and service demands
also vary.
• Customers are participants in the service and the
level of congestion impacts on perceived quality.
• Inability to control demand results in capacity
measured in terms of inputs (e.g. number of hotel
rooms rather than guest nights).
 Strategic Role of Capacity Decisions
• Using long range capacity as a preemptive strike
where market is too small for two competitors (e.g.
building a luxury hotel in a mid-sized city)
• Lack of short-term capacity planning can generate
customers for competition (e.g. restaurant staffing)
• Capacity decisions balance costs of lost sales if
capacity is inadequate against operating losses if
demand does not reach expectations.
• Strategy of building ahead of demand is often taken
to avoid losing customers.
 Queuing System Cost Tradeoff
Let: Cw = Cost of one customer waiting in
queue for an hour
Cs = Hourly cost per server
C = Number of servers
Total Cost/hour = Hourly Service Cost +
Hourly Customer Waiting Cost
Total Cost/hour = Cs C + Cw Lq
Note: Only consider systems where

C

 Queuing Formulas
Single Server Model with Poisson Arrival and Service Rates: M/M/1

1. Mean arrival rate: 

2. Mean service rate: 

3. Mean number in service:  
 n
4. Probability of exactly “n” customers in the system: Pn   (1   )
5. Probability of “k” or more customers in the system: P (n  k )   k
6. Mean number of customers in the system: 
Ls 

7. Mean number of customers in queue: 
Lq 
1 
8. Mean time in system: Ws 

9. Mean time in queue: 
Wq 

Queuing Formulas (cont.)
Single Server General Service Distribution Model: M/G/1
 2  2 2
Lq 
2(1   )

Mean number of customers in queue for two servers: M/M/2

3
Lq 
4  2
Relationships among system characteristics:
Ls  Lq  
1
Ws  Wq 

1
Ws  Ls

1
Wq  Lq

 Capacity Analysis of Robot
Maintenance and Repair Service
A production line is dependent upon the use of assembly robots that
periodically break down and require service. The average time between
breakdowns is three days with a Poisson arrival rate. The average time to
repair a robot is two days with exponential distribution. One mechanic repairs
the robots in the order in which they fail.
1. What is the average number of robots out of service?
2. If management wants 95% assurance that robots out of service will not cause
the production line to shut down due to lack of working robots, how many
spare robots need to be purchased?
3. Management is considering a preventive maintenance (PM) program at a
daily cost of $100 which will extend the average breakdowns to six days.
Would you recommend this program if the cost of having a robot out of
service is $500 per day? Assume PM is accomplished while the robots are in
service.
4. If mechanics are paid $100 per day and the PM program is in effect, should
another mechanic be hired? Consider daily cost of mechanics and idle robots.
Single Server General Service
Distribution Model : M/G/1
 2  2 2
Lq 
2(1   )
1
1. For Exponential Distribution: 2 
2
 2  2 /  2 2 2 2
Lq   
2(1   ) 2(1   ) (1   )

2. For Constant Service Time: 2  0

2
Lq 
2(1   )

3. Conclusion:

Congestion measured by Lq is accounted for equally by

variability in arrivals and service times.
General Queuing Observations
1. Variability in arrivals and service times contribute equally to
congestion as measured by Lq.

2. Service capacity must exceed demand.

3. Servers must be idle some of the time.

4. Single queue preferred to multiple queue unless jockeying

is permitted.

5. Large single server (team) preferred to multiple-servers if

minimizing mean time in system, WS.

6. Multiple-servers preferred to single large server (team) if

minimizing mean time in queue, WQ.
Managing Waiting Lines
 Lines and Waiting

“Every day I get in the queue, that waits

for the bus that takes me to you …”
Pete Townshend, Magic Bus
 Where the Time Goes
In a life time, the average
American will spend--

SIX MONTHS
Waiting at stoplights
EIGHT MONTHS
Opening junk mail
ONE YEAR
Looking for misplaced 0bjects
Unsuccessfully returning
TWO YEARS phone calls

FOUR YEARS Doing housework

FIVE YEARS Waiting in line

SIX YEARS Eating

 Cultural Attitudes
• “Americans hate to wait. So business is
trying a trick or two to make lines seem
shorter…” The New York Times, September 25, 1988
• “An Englishman, even when he is by
himself, will form an orderly queue of
one…” George Mikes, “How to be an Alien”
• “In the Soviet Union, waiting lines were
used as a rationing device…” Hedrick Smith, “The
Russians”
 Waiting Realities
• Inevitability of Waiting: Waiting results
from variations in arrival rates and service
rates

• Economics of Waiting: High utilization

purchased at the price of customer waiting.
Make waiting productive (salad bar) or
profitable (drinking bar).
 Laws of Service
• Maister’s First Law:
Customers compare expectations with
perceptions.
• Maister’s Second Law:
Is hard to play catch-up ball.
• Skinner’s Law:
The other line always moves faster.
• Jenkin’s Corollary:
However, when you switch to another other line,
the line you left moves faster.
 Remember Me
• I am the person who goes into a restaurant, sits
down, and patiently waits while the wait-staff does
everything but take my order.
• I am the person that waits in line for the clerk to
finish chatting with his buddy.
• I am the one who never comes back and it amuses
me to see money spent to get me back.
• I was there in the first place, all you had to do was
show me some courtesy and service.
The Customer
 Psychology of Waiting
• That Old Empty Feeling: Unoccupied time goes slowly
• A Foot in the Door: Pre-service waits seem longer that
in-service waits
• The Light at the End of the Tunnel: Reduce anxiety
with attention
• Excuse Me, But I Was First: Social justice with FCFS
queue discipline
• They Also Serve, Who Sit and Wait: Avoids idle service
capacity
Approaches to Controlling Customer Waiting

• Animate: Disneyland distractions, elevator

mirror, recorded music
• Discriminate: Avis frequent renter
treatment (out of sight)
• Automate: Use computer scripts to address
75% of questions
• Obfuscate: Disneyland staged waits (e.g.
House of Horrors)
 The Art of Service Recovery
“To err is human; to recover, divine”
• Measure Cost of Lost Customer
• Listen Carefully
• Anticipate Need for Recovery
• Act Fast
• Train Employees
• Empower the Frontline
• Inform Customers of Improvement
Essential Features of Queuing Systems

Renege

Arrival Queue
process Departure
Calling Queue discipline Service
population configuration process
Balk No future
need for
service
 Arrival Process
Arrival
process

Static Dynamic

Random Random arrival Customer-

Facility- exercised
arrivals with rate varying
controlled control
constant rate with time

Accept/Reject Price Appointments Reneging Balking

Distribution of Patient Interarrival Times

40
Relative frequency, %

30
20
10
0
1 3 5 7 9 11 13 15 17 19
Patient interarrival time, minutes
Temporal Variation in Arrival Rates

3.5 140
3 130
2.5 120

Percentageof averagedaily
2 110

physicianvisits
100
veragecallsperhour

1.5
90
1
80
0.5 70
0 60
A

1 3 5 7 9 11 13 15 17 19 21 23 1 2 3 4 5
Hour of day Day of week
 Poisson and Exponential Equivalence

Poisson distribution for number of arrivals per hour (top view)

One-hour
1 2 0 1 interval
Arrival Arrivals Arrivals Arrival

62 min.
40 min.
123 min.

Exponential distribution of time between arrivals in minutes (bottom view)

 Queue Configurations
Multiple Queue Single queue

Take a Number
Enter
3 4 2

8 6 10
12 7
11 9
5
 Queue Discipline
Queue
discipline

Static
Dynamic
(FCFS rule)

selection Selection based

based on status on individual
of queue customer
attributes

Number of Processing time

customers Round robin Priority Preemptive of customers
waiting (SPT rule)
 Outpatient Service Process
Distributions

15 15
10 10
frequency. %

Relativefrequency,
Relative

%
5 5
0 0
1 11 21 31 41 1 11 21 31 41
Minutes Minutes
15

10
Relativefrequency,
%

0
1 11 21 31 41
Minutes
 Service Facility Arrangements

Service facility Server arrangement

Parking lot Self-serve

Cafeteria Servers in series

Toll booths Servers in parallel

Supermarket Self-serve, first stage; parallel servers, second stage

Hospital Many service centers in parallel and series, not all used by each
patient
Strategies for Matching Supply
and Demand for Services

DEMAND SUPPLY
STRATEGIES STRATEGIES

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

Yield
management
 Segmenting Demand at a Health
Clinic

140
130 Smoothing Demand by Appointment
120 Scheduling

110 Day Appointments

Percentageof averagedaily
physicianvisits

100
Monday 84
90
Tuesday 89
80 Wednesday 124
70 Thursday 129
Friday 114
60
1 2 3 4 5
Day of week
 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
 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
 Daily Scheduling of
Telephone Operator Workshifts

2500
30
Topline profile
2000 25
20

Number of operators
1500
Scheduler program assigns
Calls

15 tours so that the number of

1000 operators present each
10 half hour adds up to the
number required
500
5 Tour

0 0
12 2 4 6 8 10 12 2 4 6 8 10 12 12 2 4 6 8 10 12 2 4 6 8 10 12
Time Time
LP Model for Weekly Workshift Schedule with
Two Days-off Constraint
Objective function:
Minimize x1 + x2 + x3 + x4 + x5 + x6 + x7

Constraints:
Sunday x2 + x3 + x4 + x5 + x6 3
Monday x 3 + x4 + x5 + x6 + x7 6
Tuesday x1 + x4 + x5 + x6 + x7 5
Wednesday x1 + x2 + x5 + x6 + x7 6
Thursday x1 + x2 + x3 + x6 + x7 5
Friday x1 + x2 + x3 + x4 + x7 5
Saturday x1 + x2 + x3 + x4 + x5 5
xi  0 and integer

Schedule matrix, x = day off

Operator 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
 Scheduling Part-time Bank Tellers
7 5 6
Tellers required

Decreasing part-time teller demand histogram

Tellers required
0 1 2 3 4 5
5
2 3 4

4 4
3 3
2 2 1
Two Full-time Tellers 1 1 5 2
1

Fri. Mon. Wed. Thurs Tues.

0

Mon. Tues. Wed. Thurs. Fri.

Obje ctive f unction:
Minim ize x 1+ x2 +x 3 +x4 +x5 +x6 +x7

C ons tr aints:
Sunday x 2 +x3 +x4 + x 5 +x6  b1
Monday x 3+x 4 +x5 +x6 +x7  b2

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
 Yield Management
• YM is a comprehensive system to
maximize revenue for capacity-
constrained services using reservation
systems, overbooking and partitioning
demand.
 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
 by Service Class for Resort
Hotel
Percentage of capacity allocated
to different service classes

20% 20% 20%

First class 30%

30%
50% 50%
Standard
60%

50% 30%
Budget 30%
10%

Peak Shoulder Off-peak Shoulder

(30%) (20%) (40%) (10%)
Summer Fall Winter Spring

Percentage of capacity allocated to different seasons

Demand Control Chart for a Hotel
350
Expected Reservation Accumulation
300
2 standard deviation control limits
250

200
Reservations

150

100

50

0
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
 Strategies in YM
• Using reservation systems
• Overbooking
• Discount Allocations
• Traffic management
• Segmenting demand
 YM Applications
• Hotel reservations
• Airline reservations
• Logistics
• Train Reservations
• Restaurant catering software
 Problems in YM
• Alienating customers

Capacity Planning and

• Cheating by knowledgeable
Queuing
customers
UNIT 5

• Lack of systematic empowerment

• Cost and implementation time are high.

 INVENTORY
Managing Facilitating Goods
Replenishment Replenishment Replenishment Customer
order order order order

Factory Wholesaler Distributor Retailer Customer

Production
Delay Shipping Shipping Item Withdrawn
Delay Delay

Wholesaler Distributor Retailer

Inventory Inventory Inventory
 Learning Objectives
• Discuss the role of information technology in managing inventories.
• Describe the functions and costs of an inventory system.
• Determine the order quantity.
• Determine the reorder point and safety stock for inventory systems
with uncertain demand.
• Design a continuous or periodic review inventory-control system.
• Conduct an ABC analysis of inventory items.
• Determine the order quantity for the single-period inventory case.
• Describe the rationale behind the retail discounting model.
 Role of Inventory in Services
• Decoupling inventories
• Seasonal inventories
• Speculative inventories
• Cyclical inventories
• In-transit inventories
• Safety stocks
 Considerations in Inventory
Systems
• Type of customer demand

• Planning time horizon

• Replenishment lead time

• Constraints and relevant costs

 Relevant Inventory Costs
• Ordering costs

• Receiving and inspections costs

• Holding or carrying costs

• Shortage costs
 Inventory Management
Questions
• What should be the order quantity (Q)?
• When should an order be placed, called a
reorder point (ROP)?
• How much safety stock (SS) should be
maintained?
 Inventory Models
• Economic Order Quantity (EOQ)
• Special Inventory Models
With Quantity Discounts
Planned Shortages
• Demand Uncertainty - Safety Stocks
• Inventory Control Systems
Continuous-Review (Q,r)
Periodic-Review (order-up-to)
• Single Period Inventory Model
Inventory Levels For EOQ
Model
Units on Hand

0
Q Time
D
Annual Costs For EOQ Model

900
800
700
Annual Cost, $

600
Holding Cost
500
Ordering Cost
400
Total Cost
300
200
100
0

Order Quantity, Q
 EOQ Formula
• Notation
D = demand in units per year
H = holding cost in dollars/unit/year
S = cost of placing an order in dollars
Q = order quantity in units
• Total Annual Cost for Purchase Lots

• EOQ TCp  S ( D / Q)  H (Q / 2)

2 DS
EOQ 
H
 Annual Costs for Quantity
Discount Model
22,000
C = $20.00 C = $19.50 C = $18.75

21000
Annual Cost, $

20000

2000

1000

0 100 200 300 400 500 600 700

Order quantity, Q
Inventory Levels For Planned
Shortages Model
Q-K

0 TIME

-K
T1 T2
T
 Formulas for Special Models
• Quantity Discount Total Cost Model

TCqd  CD  S ( D / Q)  I (CQ / 2)
• Model with Planned Shortages

D (Q  K ) 2 K2
TCb  S  H B
Q 2Q 2Q

* 2 DS  H  B 
Q   
H  B 

*  H 
*
K Q  
 H  B
 Values for Q* and K* as A
Function of Backorder Cost
B Q* K* Inventory Levels

B  2DS 0
0
H

2DS  H  B   H 
0 B   Q* 0
H  B   H  B 

B 0 undefined Q* 0
 Demand During Lead Time
Example

L  3

  15
.   15
.   15
.   15
.

+ + + =

u=3 u=3 u=3 u=3 

d L  12 ROP
ss

Four Days Lead Time Demand During Lead time

 Safety Stock (SS)
• Demand During Lead Time (LT) has
Normal Distribution with
-
Mean(d L )   ( LT )
-
• SS Std
with. Dev  L )  level
r% .(service LT

SS  z 
• Reorder Point
r LT

ROP  SS  d L
 Continuous Review System
(Q,r)
Amount used during first lead time
Inventory on hand

Order quantity, EOQ

EOQ
Reorder point, ROP

d3
Average lead time usage, dL
d1
d2 EOQ

Safety stock, SS First lead

time, LT1 LT2 LT3

Time
Order 1 placed Order 2 placed Order 3 placed

Shipment 1 received Shipment 2 received Shipment 3 received

 Periodic Review System
(order-up-to)
Inventory on Hand
Review period

RP RP RP
Target inventory level, TIL

First order quantity, Q1

Q2 Q3

d1 d3
Amount used during
first lead time

d2

Safety stock, SS First lead time, LT1

LT2 LT3

Time
Order 1 placed Order 2 placed Order 3 placed

Shipment 1 received Shipment 2 received Shipment 3 received

 Inventory Control Systems
• Continuous Review System
2 DS
EOQ 
H
ROP  SS   LT
SS  zr LT

• Periodic Review System

RP  EOQ / 
TIL  SS   ( RP  LT )
SS  zr RP  LT
ABC Classification of Inventory
Items

110
100
90
80
Percentageof dollarvolume

70
60
50
40
30
20 A B C
10
0

Percentage of inventory items (SKUs)

 Inventory Items Listed in
Descending Order of Dollar Volume
Monthly Percent of
Unit cost Sales Dollar Dollar Percent of
Inventory Item ($) (units) Volume ($) Volume SKUs Class

Computers 3000 50 150,000 74 20 A

Entertainment center 2500 30 75,000

Television sets 400 60 24,000

Refrigerators 1000 15 15,000 16 30 B
Monitors 200 50 10,000

Stereos 150 60 9,000

Cameras 200 40 8,000
Software 50 100 5,000 10 50 C
Computer disks 5 1000 5,000
CDs 20 200 4,000

Totals 305,000 100 100

Single Period Inventory Model
Newsvendor Problem Example
D = newspapers demanded
p(D) = probability of demand
Q = newspapers stocked
P = selling price of newspaper, $10
C = cost of newspaper, $4
S = salvage value of newspaper, $2
Cu = unit contribution: P-C = $6
Co = unit loss: C-S = $2
 Single Period Inventory Model
Expected Value Analysis
Stock Q
p(D) D 6 7 8 9 10

.028 2 4 2 0 -2 -4
.055 3 12 10 8 6 4
.083 4 20 18 16 14 12
.111 5 28 26 24 22 20
.139 6 36 34 32 30 28
.167 7 36 42 40 38 36
.139 8 36 42 48 46 44
.111 9 36 42 48 54 52
.083 10 36 42 48 54 60
.055 11 36 42 48 54 60
.028 12 36 42 48 54 60

Expected Profit $31.54 $34.43 $35.77 $35.99 $35.33

 Single Period Inventory Model
Incremental Analysis

E (revenue on last sale)  E (loss on last sale)

P ( revenue) (unit revenue)  P (loss) (unit loss)

P( D  Q)Cu  P( D  Q)Co

1 P( D  Q)C u  P ( D  Q)Co

Cu
P ( D  Q)  (Critical Fractile)
Cu  Co
where:
Cu = unit contribution from newspaper sale ( opportunity cost of underestimating demand)
Co = unit loss from not selling newspaper (cost of overestimating demand)
D = demand
Q = newspaper stocked
Critical fractile for the
newsvendor problem
P(D<Q)
(Co applies)

P(D>Q)
(Cu applies)
Probability

0.722

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14
Newspaper demand, Q
 Retail Discounting Model
• S = current selling price
• D = discount price
• P = profit margin on cost (% markup as decimal)
• Y = average number of years to sell entire stock of “dogs” at
current price (total years to clear stock divided by 2)
• N = inventory turns (number of times stock turns in one year)

Loss per item = Gain from revenue

S – D = D(PNY)

S
D
(1  PNY )