Analysis and Management

of Production System
Lesson 2: Introduction to Factory Models

Prof. Giulia Bruno

Department of Management and

Production Engineering

[email protected]
Factory model

● Analytical approach for the modeling and analysis

of manufacturing and production systems
 Firstly, extremely simple models (single machine models) are
developed to show the necessary mechanics and concepts
needed
 Then, more complex models are developed by connecting
simple models into networks of workstations with the appropriate
interconnections
● Overall approach
1. decompose a system into small components
2. model components
3. reintegrate the general system by appropriate combination of
submodels
Factory model

● Decomposition approach is an approximation

 Produces acceptable results in a wide variety of manufacturing
applications
● Any analytical model (exact or approximate) is an
approximation of the real world environment
● Question to be answered: whether or not the model
yields accurate enough results to be used as an
analysis tool in support of design and operational
decision making
Factory model

● Modelled system doesn’t take into account where or why

jobs arrive or how they are transported to customers
 modeling the order creation or completed job delivery systems is not
within the scope of our analysis
● Arriving job is a physical entity to be processed through the
various processing steps or an order to begin the processing
of raw material into a newly manufactured entity
Notations and definitions

● Factory composed by several machines grouped together by

type (called workstations) and series of jobs to be produced
on the machines
● Processing steps for a job consists of several processing
operations to be performed by different machines in a
specified sequence
● Jobs move through the factory, waiting in line at a
workstation until their turn for being processed by the
machine, then proceeding to the next workstation to repeat
this sequence until all required operations have been
completed
● Jobs arrive at the factory either individually or in batches
based on some distribution of the time between arrivals
Workstation

• Collection of one or more identical machines or

resources (i.e. machines which perform the same
function)
Routing

● Sequence of processing steps for a job

● Jobs with identical routings are said to be of the same
job type
 different job types are jobs with different routings
Performance measures

• Cycle time: time a job spends in a system

• Average cycle time of the line is denoted as CT
• Sometimes called Lead Time, even if LT identifies the a priori
evaluation of the cycle time (time allocated to a operation
sequence to produce a certain job)
• Work in process: number of jobs within a system (from the
beginning to the end of the operations sequence) which are
either undergoing processing or waiting in a queue for
processing
• Average work-in-process is denoted as WIP
• Throughput rate: number of completed jobs leaving the
system per unit of time
• Throughput rate averaged over many jobs is denoted by TH
Cycle time

• Notational distinction
• average factory cycle time denoted as CTs
• average cycle time at workstation I, denoted as CT(i)
• CTs: average time that a job spends within the factory, either
being processed at a workstation or waiting in a workstation
queue;
• CT(i): average time jobs spend being processed by
workstation i plus the average time spend in the workstation i
queue (or buffer)
• CT: used for general properties related to the average cycle
time will be developed
Cycle time

• Processing time often known or can be determined without

much effort
• Queue time not easily estimated for a given job
• it depends on the number and processing times of the various types of
jobs that are waiting in the queue ahead of the designated job
• Average cycle time at workstation i given as the sum of two
components:

CT(i) =CTq(i)+E[Ts(i)] ,

where Ts(i) denotes the service time (or processing time) at

workstation i
Factory analysis

● For most analyzed systems, long-run throughput

rate is equal to the input rate of jobs
● Main issue: estimation of the total length of time
for the manufacturing process (CTs)
● The higher the factory capacity, the faster jobs
are completed
 Cycle time increases as the factory becomes busier
Factory analysis

● Time dependent measures such as CTs(t) and WIPs(t)

are very difficult to develop
 Focus restricted to the “steady-state” measures i.e., the
limiting value of the time dependent measures
 By a property called the ergodic property, steady-state
values can also be considered to be time-averaged values
as time becomes very large
● Steady-state measures are independent of the initial
conditions of the system
● In the queueing theory, results are for steady-state system
measures
Factory analysis

● System’s performance measures CT and WIP can be

estimated from the arrival and departure streams of the
system
● Define Tai as the arrival time of the ith job, and the function
A(t) for t ≥ 0 as the total number of arrivals during the
timeinterval [0, t]
● Define Tdi as the departure time of the ith job, and the
function D(t) for t ≥0 as the total number of departures
during the interval [0, t]
Factory analysis

● Consider a time interval (a, b) such that the system starts

empty and returns to empty
● Let Nab be the number of jobs that arrive to the system
during the interval (a, b)
● Number these jobs from 1 to N, with index i representing
specific jobs.
● Then the average waiting time, CT(a,b), for jobs during
this interval is given by
Factory analysis

● The area, AB, between the curves A(t) and D(t) for a < t <
b is merely the summation given in the above equation.
This area can also be obtained by standard integration
methods as

● The area represents the integral of the number of jobs in

the system at time t, since N(t) = A(t)−D(t) is the number
of jobs in the system at t. So the time-averaged number of
jobs waiting in the system during the time interval (a, b) is
given by
Factory analysis

● There is a relationship between the average number in

the system during the interval (a, b) and the average
waiting time or cycle time in the system during this
interval. Since the area between A(・) and D(・) (namely
AB) is constant regardless of the method used to measure
it, we have

● Thus, the following relationship is obtained

Factory analysis

● The mean number of jobs arriving to the system per unit

time, normally denoted as λ, is Nab/(b−a), thus

● This result is valid for any interval that starts with an

empty system and ends with an empty system
● In fact this relationship is the limiting behavior result, or
long run average result, for stationary queueing systems,
and is known as Little’s Law
● The result holds for individual workstations as well as the
system as a whole
Factory analysis

● For a system that satisfies steady-state conditions, the

following equation holds

WIP =λ×CT

where WIP is the long-run average number of jobs in the

system, CT is the long-run average cycle time and λ is the
long-run input rate of jobs to the server

Since the average input rate is usually equal to the average

throughput rate, Little’s Law can also be written as
WIP = th×CT
Factory analysis

● The formula estimates mean values, but the actual

underlying random variables of the systems can be quite
variable
● In most single workstation system models, the average
number in the system, WIP, can be easily obtained.
However, the behavior of the random variable
representing the number in the system at any one point in
time can be highly variable
Factory analysis

● Term steady-state implies that the mean

reaches a limiting value and thus ceases to
change with respect to time
 It does not imply that the system itself ceases to change
● Variability continues forever
 fluctuations within the system never cease
● Steady-state does imply that the entire
distribution reaches a limiting value
 not only the mean but also the standard deviation,
skewness, and other such measures will have limiting values
Exercise 1

A workstation with a single machine for processing

has a long-run average inventory level (WIP) of 25
jobs. The average rate at which jobs enter the
workstation is 4 per hour, and the average
processing time is 14.5 minutes per job. What is
the average time that a job spends in the queue?
Exercise 1 – Solution

DATA
• WIP = 25 pcs
• rate = λ = 4 pcs/hr
14.5
• E[T𝑆] = 14.5 min = = 0.24 hrs
60
• CTq = ?

λ TH = λ
WS
Exercise 1 – Solution

According to the Little’s Law: 𝑊𝐼𝑃 = 𝑇𝐻 × 𝐶𝑇

Supposing to have no losses or reworks, we can say
that 𝑇𝐻 = 𝜆 and the equations becomes:
𝑊𝐼𝑃 = 𝜆 × 𝐶𝑇
Thanks to this, we can find out the Cycle Time of the
station:
𝑊𝐼𝑃 25
𝐶𝑇 = = = 6.25 ℎ𝑟𝑠
𝜆 4
We also know that 𝐶𝑇 = 𝐶𝑇𝑞 + 𝐸[𝑇𝑆]
So,
𝐶𝑇𝑞 = 𝐶𝑇 − 𝐸[𝑇𝑆 ] = 6.25 − 0.24 = 6.01 ℎ𝑟𝑠
Exercise 2

Consider a factory operating 24 hours per day

consisting of two workstations. Arrivals to the first
station occur at a rate of 10 per day. The long-run
average time that a job spends at the first workstation
is 4.2 hours. After processing at the first workstation,
a job is sent directly to the second workstation where
it spends an average of 5.3 hours. After processing at
the second workstation, the job leaves the system.
What is the average work-in-process within the
factory?
Exercise 2 – Solution

• 𝑆ℎ𝑖𝑓𝑡 = 24 ℎ𝑟𝑠/𝑑𝑎𝑦
10
• 𝑟𝑎𝑡𝑒 (𝜆) = 10 𝑝𝑐𝑠/𝑑𝑎𝑦 = 𝑝𝑐𝑠/ℎ𝑟 = 0.417 𝑝𝑐𝑠/ℎ𝑟
24
• 𝐶𝑇1 = 4.2 ℎ𝑟𝑠
• 𝐶𝑇2 = 5.3 ℎ𝑟𝑠

WS 1 WS 2
Exercise 2 – Solution

Computation of WIP level related to WS1 and

WS2 with the Little’s Law:
● 𝑊𝐼𝑃1 = 𝜆 × 𝐶𝑇1 = 0.417 × 4.2 = 1.75 𝑝𝑐𝑠
● 𝑊𝐼𝑃2 = 𝜆 × 𝐶𝑇2 = 0.417 × 5.3 = 2.21 𝑝𝑐𝑠

Total WIP given by the sum of the two WIP

related to the single stations:
● 𝑊𝐼𝑃𝑡𝑜𝑡 = 𝑊𝐼𝑃1 + 𝑊𝐼𝑃2 = 1.75 + 2.21 = 3.96 𝑝𝑐𝑠
Penny Fab model
● Factory that makes only one type of product
● Four processing steps to be performed in sequence
● Each processing operation performed on a separate machine
● Machines process only one unit of the product at a time (a job)
● Constant processing times: 1, 2, 1 and 1 hour(s) respectively
● Constant WIPs
 when a job has completed its four processing steps, it is immediately
removed from the factory and a new job is started at Machine 1 to keep
the total factory WIPs at the specified level
Penny Fab model

● Constant WIPs level fixed at 10 jobs decided by the

Management
● Result a th = 0.5 job/h (one finished job every two hours on
the average is produced)
● Equal to the maximum throughput rate for this factory because
its slowest processing step (at Machine 2) takes two hours per
job
● Jobs can be completed no faster than this single machine
completes its own processing
Penny Fab model

● Management pleased with the throughput of the factory since

it is at its maximum capacity, but concerned with the cycle
time (20 hours per job)
 very high since it only takes 5 hours of processing to complete each job
● The x-factor for a factory is the ratio of CTs to the average
total processing time per job
 X-factor = 4
● From literature analysis the x-factor for this sector is 2.6
 management is worried about the ability to keep customers when
the industry on average produces the same product with a
considerably shorter lead-time from order placement to receipt
Penny Fab model

● Possible solution to address the cycle time problem: purchase

a 25% faster machine (1.5 hours) for processing step two
● This purchase would be made expressly for the purpose of
reducing the x-factor for the factory to be more in line with the
industry average
● The company selling the machine says that this investment
will bring the x-factor down to 3.33 and the additional
throughput of 0.166 units per hour would pay for the cost of
the new machine in three years
Penny Fab model

● Management decided that investment is not worthwhile just

based on increased throughput
 funds needed to buy the machine are needed for other aspects of
the company (e.g., research and development)
● Hiring of a consulting team from the manufacturing
engineering department of a local university to perform a short
term factory flow analysis study
 First activity to devise a method of predicting the long-term
factory performance measures of cycle time and throughput
Penny Fab model

● Hand simulation procedure of the factory flow

● Beginning: 10 jobs all placed at Machine 1, then hourly
updates to each job’s status
● Jobs soon distributed themselves throughout the factory and
after a short period of time a two-hour cyclic pattern emerged
● Every cycle of this pattern produced one completed job and
the factory returned to the identical state for each machine
and associated queue
 This set of conditions is referred to as the factory status
● All the job cycle times had identical values after the system
reached the cyclic behavior pattern (20 h, agreed with the
company)
Penny Fab model

● Factory simulation with

WIP = 10 for one 24-hour
day
● Status (X,Y)
 X: number of jobs at the WS
(queue + processing)
 Y: number of hours already
completed for the job in
processing
● After hour 15 the factory
status repeats every two
hours (factory status at the
start of hours 15, 17, 19,
21, etc. are identical)
Penny Fab model

● The consulting team decided to estimate the x-factor for

various numbers of jobs in the system (WIP)
● From Little’s Law: CT=WIP/th
● In this case, th=0.5 j/h, then CT=2*WIP
● Processing time = 5h, then x=CT/5=WIP/2.5
● In order to have the desired x-factor (2.6), the WIP must be
6.5
● How the th change by varying WIP?
Penny Fab manual simulation

WS1 WS2 WS3 WS4

WIP = 1
t1=1h t2=2h t3=1h t4=1h

t = 0h WS1 WS2 WS3 WS4

1 item every 5
t = 1h WS1 WS2 WS3 WS4 hours comes out
of the line (TH =
t = 2h WS1 WS2 WS3 WS4 1/5 pcs/h)

t = 3h WS1 WS2 WS3 WS4 Each item stays

on the line for 5
t = 4h WS1 WS2 WS3 WS4 hours (CT = 5 h)
t = 5h WS1 WS2 WS3 WS4

WIP=1 CT=5h TH=1/5=0.2 pz/h

Penny Fab manual simulation

WS1 WS2 WS3 WS4

WIP = 2
t1=1h t2=2h t3=1h t4=1h

t = 0h WS1 WS2 WS3 WS4

2 item every 5
t = 1h WS1 WS2 WS3 WS4 hours comes out
of the line (TH =
t = 2h WS1 WS2 WS3 WS4 2/5 pcs/h)

t = 3h WS1 WS2 WS3 WS4 Each item stays

on the line for 5
t = 4h WS1 WS2 WS3 WS4 hours (CT = 5 h)
t = 5h WS1 WS2 WS3 WS4

WIP=2 CT=5h TH=2/5=0.4 pz/h

Penny Fab manual simulation
WS1 WS2 WS3 WS4
WIP = 3
t1=1h t2=2h t3=1h t4=1h

t = 0h WS1 WS2 WS3 WS4

3 item every 6
t = 1h WS1 WS2 WS3 WS4 hours comes out
of the line (TH =
t = 2h WS1 WS2 WS3 WS4 3/6 pcs/h)

t = 3h WS1 WS2 WS3 WS4 Each item stays

on the line for 6
t = 4h WS1 WS2 WS3 WS4 hours (CT = 6 h)
t = 5h WS1 WS2 WS3 WS4
t = 6h WS1 WS2 WS3 WS4
WIP=3 CT=6h TH=3/6=0.5 pz/h
Penny Fab manual simulation
Penny Fab manual simulation

● The WIP level in the factory can be reduced to 3 jobs

while maintaining the factory throughput rate of ½
● The cycle time reduces to 2×WIP = 6 hours with an x-
factor of 1.2
● Thus at no expense, the factory can maintain its current
throughput rate and reduce its cycle time from 20 to 6
hours
Conclusion

● CT remains constant at the minimum CT (T0) as WIP

increases and it increases to infinity once the maximum
WIP (rb) is reached
● TH increases as WIP level increases, until the WIP level
reaches the maximum WIP (rb); after that TH stabilizes,
remaining constant
● The best case (min CT and max TH) is when with WIP = 3
Line critical parameters

1. Bottleneck Rate (rb): TH of the workstation which is

the bottleneck of a line (maximum TH obtained from
the line)
2. Raw process time (T0): sum of the long-term
average process times of each station in the line
(minimum possible CT)
3. Critical WIP (W0 = rb ∙T0): WIP level which allows a
line to achieve the maximum TH (rb) with the
minimum cycle time (T0)
Example: balanced production line

• Line composed of 4 machines in series

• Each machine has a 2 hours per piece process time
• Capacity of each machine (inverse of process time) =
1 pcs/2h
• Once an operation on a machine is completed, the
product goes immediately to the next machine
• Machines work continuously, with no stops
• No discarded pieces
Example: balanced production line

Critical parameters:
• rb = 1 pcs/2h = 0.5 pcs/g
(max utilization ≡ min capacity)
• T0 = 2h*4 = 8h
(sum of process times on machines)
• W0 = rb*T0 = 0.5*8 = 4 pcs

W0 is the critical value of WIP, which allows to reach

rb = 0.5 pcs/h and T0 = 8 h
Example: not balanced production line

• Line composed of work stations in series

• Each WS is composed of a different number of machines
• Each WS is composed of identical machines (with the same
process time) in parallel
• WS capacity = single machine capacity * number of machines
Machine
capacity WS capacity
Work Station Machines Process time
(inverse of (pcs/h)
process time)

1 1 2

2 2 5

3 6 10

4 2 3
Example: not balanced production line

• Line composed of work stations in series

• Each WS is composed of a different number of machines
• Each WS is composed of identical machines (with the same
process time) in parallel
• WS capacity = single machine capacity * number of machines

Machine
capacity WS capacity
Work Station Machines Process time
(inverse of (pcs/h)
process time)

1 1 2 ½ = 0.5 1*0.5 = 0.5

2 2 5 1/5 = 0.2 2*0.2 = 0.4

3 6 10 1/10 = 0.1 6*0.1 = 0.6

4 2 3 1/3 = 0.33 2*0.33 = 0.67

Example: not balanced production line

Critical parameters:
• rb = 0.4 pcs/h (WS2 capacity: not WS3 with
slower machines and not WS1 with smaller
number of machines)
• T0 = 20 h (sum of process times)
• W0 = 0.4*20 = 8 pcs (it is smaller than the
number of machines, because some station
won’t be completely used)
HAL – Large Panel Line Processes

Data:
• Average process rate (number of panels/h)
• Average process time (h) at each station

Batches and parallel machines are present, so rate is different

from 1/time
HAL – Large Panel Line Processes

● Large Panel Line: produces printed circuit boards

● Line runs 24 hr/day (only 19.5 hrs of productive time)
● Recent Performance (last 7 months):
 TH = 1,400 panels per day (71.8 panels/hr)
 WIP = 47,600 panels
 CT = 34 days (663 hr at 19.5 hr/day)
 Customer service = 75% on-time delivery

How HAL is performing?

Which data we need to evaluate?
HAL – Large Panel Line Processes

• Critical WIP: rbT0 = 144 x 33.9 = 3,869

• Actual Values:
 CT = 34 days = 663 hours (at 19.5 hr/day)
 WIP = 47,600 panels
 TH = 71.8 panels/hour
• Conclusions:
 TH is 63% of capacity
 WIP is 12.3 times critical WIP
 CT is 24.1 times raw process time
 The factory is not performing well!
Exercise

Consider a production line composed of four

workstations, each consisting of a single machine,
except the second station which consists of two identical
machines in parallel
Each machine has mean process time equal to two
hours per job
a) Determine the bottleneck rate rb, the raw process
time T0 and the critical WIP of the line
b) Compute CT and TH for values of WIP going from 1
to 10
Exercise – Solution
Exercise 2

Consider a production line composed by three

workstations in series. The characteristics of the
workstations are reported in the table.
Compute the bottleneck rate rb, the raw process time T0
and the critical WIP W0 of the system.
What happen to the TH and CT values when the WIP
exceeds W0?

Workstation i E[Tsi] (min) Machines

1 E[Ts1] 3 2
2 E[Ts2] 2 1
3 E[Ts3] 4 3
Solution

min
j/min
j
Solution

CT TH

If the WIP exceeds W0, the TH remains constant at value

of 0.5 j/min and the CT increases at a rate of WIP/0.5