ISOM 2700: Operations Management

Session 2: Process Analysis I

Dongwook Shin
Dept. ISOM, HKUST Business School
Course Roadmap
Bottleneck
Little’s law
Utilization

Maximize
Profits

1
Learning Objectives: Session 2

• Basic Principles of Process Analysis

• Little’s Law: Inventory, Flow Rate, and Flow Time

• Utilization and Bottleneck

2
The Process View (Focus of This Class)

Inputs Outputs
Transformation
Process

5
Inside the Black Box…

Inputs Transformation
Outputs
Process
Labor Products
Facility &
Energy Services
Material
Information

Process is a network of activities that

transform inputs into outputs

6
 Process Flow Chart: Terminology
Flow rate
Demand rate
Prep Cook Pack (throughput
(input rate)
rate)
10 units/hour 5 minutes 6 minutes 2 minutes

Activity: Carried out by resources, Have a capacity

Inventory/Buffer: Do not have a capacity

Flow of goods/materials

7
 Measures of Process Performance
Flow rate
Demand rate
Prep Cook Pack (throughput
(input rate)
rate)
10 units/hour 5 minutes 6 minutes 2 minutes

• Activity time: how long does the task spend on the activity?
• Capacity = 1 / activity time: how many units can be made per unit of time
• Bottleneck: a resource with the smallest capacity
• Process capacity: capacity of the bottleneck
• How much the process can produce?
• Flow rate = Minimum{Demand rate, Process capacity}
• How much the process actually does produce?
• Cycle time = 1 / flow rate: the time between two successive product completions
• Flow time: the time a unit spends in the process (time worked on + time in buffer)
• Inventory: the number of flow units contained within the process
8
Capacity & Demand Constraints
Flow rate = Minimum{Demand rate, Process capacity}

Bottleneck Demand
Input

Capacity- Flow Rate

Constrained

Bottleneck
Input

Demand
Demand-
Constrained Flow Rate

9
Gantt Chart: Another Perspective…
Time (in minutes)
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25
Prep
Cook
Wrap

• It takes 13 minutes to product 1 unit, i.e., flow time = 13 minutes

• The time between the two successive outputs is 6 minutes, i.e., cycle time = 6
minutes
• The process can produce 10 units per hour, i.e., process capacity = 10 units
per hour
• Inventory varies through time
• Inventory = 2 units at time 7
• Inventory = 3 units at time 13

10
Learning Objectives: Session 2

• Basic Principles of Process Analysis

• Little’s Law: Inventory, Flow Rate, and Flow Time

• Utilization and Bottleneck

11
 Why Should We Care Inventory?

... ... Prep Cook Pack ... ...

• Accountants view inventory as an asset

• In common speech, inventory can work to one’s
disadvantage, which can clearly be a liability
• In general, too many or too little inventory can be costly

• How to measure inventory? (Today)

• In the Inventory Management Module later this course, we will
study how to control inventory
12
Little’s Law: A Motivation
3rd in 4th in 5th in
2nd in 1st out 2nd out 3rd out

1st in

3
Inventory in
2
the system Flow time
1
Time (in minutes)
0 6 12 13 18 19 24 25 30 31

• Brute force approach to compute average inventory

• Count the inventory at every moment in time and take the
average, yielding average inventory = 2.2 units

13
What if the Process is More Complicated?
The Process Flow Chart of National Cranberry
(Bins 1-16) 3*(1500 bbls/hr)
Destoning
Dry = 4500 bbls/hr
(3)
storage
Dry
Kiwanee
Dry
Dumpers
3*(400bbls/hr) 3*(400 bbls/hr)
(5)
Dry = 1200 bbls/hr = 1200 bbls/hr
4000 bbls

y
Truck Dr
Queue (Bins 17-24) Dry Separator Bailey
Dry Storage Dechaffing Lines nd Mills
2
(3) (3) (3)
Quality
Berries
Weigh Wet Wet Dryer
Grade Wet s Wet
& Sample (3)
2000 bbls

(Bins 25-27) Wet

Wet
Wet
Storage
Avg. 75 bbls 1500 bbls/hr each
per truck allocated between the 3*(200 bbls/hr)
two types. = 600 bbls/hr
2250—4500 bbls/hr
3*(1500bbls/hr) Bad Berries
(5-10 min/truck)
= 4500bbls/hr
1200 bbls

Shipping

14
Little’s Law

I=RxT

Average Flow Rate Average

Inventory [# units/time] Flow Time
[# units] [time]

• How to remember it: units…

• Why useful?
• Out of the three fundamental performance measures (I, R, T), two can
be chosen by management, the other one is given by nature

• For fixed flow rate: Reducing inventory = reducing flow time

• Little’s law can be applied to any part of the process
15
Little’s Law: Examples
• On average how many women are pregnant in Hong Kong?
• The birth rate = 60,000 new borns / year

• Indirect measurement of flow time at Burger King

• Meat
• Flow rate = 5,000 kg / week
• Average inventory = 2,500 kg
• Flow time = ?
• Customers
• Flow rate = 1,500 customers / week
• Average inventory = 25 customers
• Flow time = ?

16
Practice Problem: Auto-Moto
Financial Services
• Auto-Moto Financial Services provides financing to qualified
buyers of new cars. Auto-Moto processes about 1,000 loan
applications per month and makes accept/reject decisions
based on an extensive review of each and every application.
Assume a 30-day working month.
• On average, only 20% of all applications received approval. An
internal audit showed that, on average, Auto-Moto had about
500 applications in process at various stages of the approval
procedure. In response to customer complaints about the time
taken to process each application, Auto-Moto decides to help
streamline its decision-making process.

17
Practice Problem: Auto-Moto
Financial Services
20% Approved 200 per month
1,000 per
Review
month
80% Rejected 800 per month

• For this process, R = 1,000 applications/month and I = 500

applications
• Average flow time?
T = 500 / 1,000 = 0.5 month (or 15 days)

18
From Diagnosis to Improvement

• The average flow time of the existing process is too long

• Suggested changes
• Set up an Initial Review Team to pre-process all applications
• Each application will then be classified into three groups:
A (high potential), B (needs more detailed evaluation), and C
(reject)
• Only A and B applications will be forwarded to different
specialists for further review and decisions

19
Findings from the Trial
Implementation
• Distribution of applications
• A: 25%
• B: 25%
• C: 50%
• About 70% of all Group A applications and 10% of Group
B applications were eventually approved
• Internal audit checks further revealed that, on average,
200 applications were with the Initial Review Team
undergoing preprocessing, 25 were with the Group A
Team undergoing the next stage of processing, and about
150 with the Group B Team
20
An Improved Process?

Group A 200/month
70%
Review Approved
IA=25
30%

25%

1,000/month Initial 25% Group B 10%

Review Review
IIR=200 IB=150
50% 90%
C 800/month
Rejected

21
Analysis for the New Process
• The flow rate and inventory for the new process:
• R = 1,000 applications per month
• I = 200 + 150 + 25
= 375 applications
• The new flow time:
• T = 375 / 1000
= 0.375 months (or 11.25 days)

• Conclusion
• There is evidence of improvement because the flow time drops
from 15 days to 11.25 days

22
Customers in IKEA

23
Customers in IKEA

• The amount of time customers spend in the store is of

interest to the management

• How would you go about finding the average time that

customers spend in the store?

24
Learning Objectives: Session 2

• Basic Principles of Process Analysis

• Little’s Law: Inventory, Flow Rate, and Flow Time

• Utilization and Bottleneck

25
Focus on the Bottleneck
• What is the bottleneck?
• The resource with the smallest capacity

• Process performance is dictated not by total activities

required, but by the bottleneck

• An analogy: Law of the Minimum

• The capacity of a barrel with staves of
unequal length is limited by the shortest
stave

Lucy Candy Video

26
Identifying Bottleneck: Circored
Plant Example
120 tons / hr 110 tons / hr
Lock First Second
Iron ore Preheater
Hopper Reactor Reactor

Finished Flash
Briquetting Discharger
good Heater
165 tons / hr 118 tons / hr 135 tons / hr

• The first reactor can hold 28 tons at maximum and the iron ore needs
to spend 15 minutes in the reactor
• The second reactor can hold 400 tons at maximum and the iron ore
needs to spend 4 hours in the reactor
27
Identifying Bottleneck: Circored
Plant Example

• The first reactor can hold 28 tons at maximum and the

iron ore needs to spend 15 minutes in the reactor
• Flow time = 15 minutes = 0.25 hours
• Inventory (at maximum) = 28 tons

• Little’s law: 28 tons = Flow rate x 0.25 hours

• Capacity = Flow rate (at maximum) = 112 tons/hour

• The capacity for the second reactor can also be computed

from Little’s law: 400 tons = Flow rate x 4 hours
• Capacity = Flow rate (at maximum) = 100 tons/hour

28
Identifying Bottleneck: Circored
Plant Example
120 tons / hr 110 tons / hr 112 tons / hr 100 tons / hr
Lock First Second
Iron ore Preheater
Hopper Reactor Reactor

Finished Flash
Briquetting Discharger
good Heater
165 tons / hr 118 tons / hr 135 tons / hr

• The bottleneck is the second reactor

• Process capacity is the capacity of bottleneck, i.e.,
process capacity = 100 tons/hr
29
Utilization

Flow Rate
• Utilization =
Capacity

• Measures how much the process actually produces

relative to how much it could produce

• The bottleneck is the resource with the highest

utilization

30
Utilization: Circored Plant Example
• Demand rate for the Circored process is 75 tons/hour
Process Step Calculations Utilization
Preheater 75/120 62.5%
Lock hoppers 75/110 68.2%
First reactor 75/112 66.9%
Second reactor 75/100 75.0%
Flash heater 75/135 55.6%
Discharger 75/118 63.6%
Briquetting 75/165 45.5%
Total process 75/100 75.0%

• Even the bottleneck might not be 100% utilized

• Is it desirable to have 100% utilization? 31

Takeaways
• First concepts of process analysis
• Fundamental performance measures: Inventory, Flow
Rate, and Flow Time
• Little’s Law relates performance measures
• Identifying the bottleneck operation is key for
complicated process
• Watch video clips to wrap up Ops in NYC Video I
Ops in NYC Video II

• Next class: More on process analysis…

32