Product Design Issues

INDR 371 Fall 2020

 Effective product design process
• Matches product characteristics with
customer requirements
• Ensures that customer requirements are met
in a cost-efficient manner
• Reduces the time to design and manufacture
 Identifying the customer needs
and the market
• Customer needs (New product ideas) from:
– R&D department
– Marketing research
– Customer complaints/suggestions
– Salespeople, Suppliers, Plant workers etc.
• Is there a market?
– Customer segments (demographics factors,
psychological factors etc.)
– Competitors (A ‘Gap Map’ may be useful)
A Determinant Gap Map (for Snacks)

S ?
O N
NTI
E
D IM
E
MOR
O R
N 4
IS I
H T: ALY
S
OUG I S AN
TH T TH
O NDU R C
F
D CO
O
FO W TO
HO

Source: Crawford and Di Beneditto

 Is there a market for this product?
• IKEA US introduction:
– plates and glasses are unsold
– Vases are selling like crazy

• American consumer prefers

– large glasses (lots of ice and large quantities of
beverage)
– Large plates (must hold 6 inch pizza)
Is there a market for this product?
Market
Research

Qualitative Quantitative
Methods Methods

Delphi
Data Analysis

Sales Force
Composite
Surveys (online,
mail, phone
Product Life- etc..)
Cycle Analogy
 Product Development Process
Concept
Concept System-Level
System-Level Detail
Detail Testing
Testingand
and Production
Production
Planning
Planning Development Design Design Refinement Ramp-Up
Development Design Design Refinement Ramp-Up

Four Phases of Product Development

The product planning phase precedes the product development process!!!

Critical Decision in Product Planning involves:

1. The portfolio of products to be developed by the organization
2. The timing of their introduction to the market (or frequency)
 Factors in Product Planning
• Market structure  Opportunities
• Risk structure  Degree of newness desired
• Resource structure Amount of breakthrough
and derivative projects
 Concept Development Process
Mission Development
Statement Identify Establish Generate Select Test Set Plan Plan
Customer Target Product Product Product Final Downstream
Needs Specifications Concepts Concept(s) Concept(s) Specifications Development

Target Specs Final Specs

Based on customer Based on selected
needs and concept, feasibility,
benchmarking models, testing, and
trade-offs

Perform Economic Analysis

Benchmark Competitive Products

Build and Test Models and Prototypes

 Product Specs
• Set Target Specifications
– Based on customer needs and benchmarks
– Develop metrics for each need
– Set ideal and acceptable values
• Refine Specifications
– Based on selected concept and feasibility testing
– Technical modeling
– Trade-offs are critical
• Reflect on the Results and the Process
– Critical for ongoing improvement
 From User Needs to Product Definition:
House of Quality
1. The customer attributes
Describing what the product must do, a structured list of needs and wants,
determined by market research. Represents the Voice of the Customer
2. The engineering characteristics
Describing how the product may achieve its required performance in general terms
which are not solution specific. Represents the Voice of the Designer.
3. Relationships
Between the customer attributes and the engineering characteristics, indicating
where there are strong, moderate or weak relationships.
4. Technical matrix
Indicating the technical priorities based on the relationships between customer
requirements and engineering characteristics. Also providing quantitative design
targets for each of the engineering characteristics, based on the technical priorities
and competitive benchmarking.
5. Technical Correlations
Recording whether the engineering characteristics mutually supporting or
contradictory
6. Planning Matrix
Providing quantitative market data for each of the customer attributes. Values can
be based on user research, competitive analysis or team assessment
 Quality Function Deployment
(House of Quality)

technical
correlations

relative engineering
importance metrics

customer benchmarking
needs on needs
relationships between
customer needs and
engineering metrics

target and final specs

 House of Quality

Source:Hauser and Clausing

 Product Architecture
Definition: The arrangement of functional elements into
physical chunks which become the building blocks for the
product or family of products.

module
module

module
module
Product
module
module

module
module
 Functional vs. Physical Elements
• Functional elements are the individual operations and transformations that
contribute to the overall performance of the product
• Physical elements are the parts, components, and subassemblies that
implement the products functions
• Example: Coffee maker

Source: Mikko Salonen

 Integral Product Architectures
• Functional elements are implemented in a very specific
manner.
• Interactions between chunks are product specific.

Compact Camera

Integral architecture increases performance and reduces

Why to choose integral architectures?
(generally) costs for any specific product model.
 Modular Product Architectures
• Each physical chunk implements one or a few functional
elements in their entirety.
• Interactions between chunks (interfaces) are well defined.

Desktop Computer
Modular architecture increases product variability.
Why to choose modular architectures?
Integral vs Modular Architectures
?
Integral Modular

How about a computer?

Some computer brand 

 Importance and Impact of Product
Architecture
Impact on the product include:
• How the product can be changed (physical changes
required to achieve a functional change)
• Product performance
• How much product variety offered
 Importance and Impact of Product
Architecture
Impact on the product design and development include:
• How functions are realized
• How component reuse and standardization is achieved
• Where module and subassembly boundaries are
• How development work is divided up
• How much product variety can be designed
 Importance and Impact of Product
Architecture
Impact on use and retirement include:
• How the product is serviced
• How the product is updated
• How the product is recycled
 Product Design Principles
• Design for Manufacturing

• Concurrent manufacturing
– Marketing, design and engineering must work
together
In order to prevent:
– Beautiful design but too expensive to
manufacture!
– Or minimal manufacturing cost but does not
appeal to consumers!
 Product Design Principles
(Continued)
• Design for logistics
– Key Concepts of Design for Logistics
• Economic packaging and transportation
• Concurrent/Parallel Processing
• Standardization
 Economic Transportation and
Storage
• Design products so that they can be
efficiently packed and stored
• Design packaging so that products can be
consolidated at cross docking points
• Design products to efficiently utilize retail
space
 Best Practices
• Ikea
– World’s largest furniture retailer
– 131 stores in 21 countries
– Large stores, centralized manufacturing,
compactly and efficiently packed products
• Rubbermaid
– Clear Classic food containers - designed to fit
14x14” Wal-Mart shelves
 Product Design Principles
• Design for product life cycle
– Manufacturing
– Supply chain
– End of Life

Ex: EU legislations force manufacturers take back

and recycle/remanufacture (some) used products.
 Good and not so good
examples…
Product design must include reverse logistics
and remanufacturing/disposal costs.
– Xerox lease / take back / remanufacture
program is a big success and led to major
savings.
– The Arcelik Telve machine is very difficult to
disassemble.
 Product Design Principles
• Design for Sustainability
– Consider the impacts of manufacturing and supply
chain processes on the environment.
• Local replenishment (may increase costs but reduces
transportation impacts)
• Avoid air shipments (longer lead times)
• Improved products / packaging

Ex: Condensed detergents take up a smaller volume and can be packed

efficiently. Decreases the transportation requirements (and the impacts)
significantly.
 Product Design

?
Quantity or
Product Machine
Production
Design Requirements
Rate

Assembly Drawing
1.Machines
(a) Exploded view diagram
2.Tools
(b) Exploded photograph
3.Equipment
Component Part Drawing
(a) Dimensions
(b) Specifications
 Process Design

P
D
Assembly Chart
Process • Sequence of operations for each part Op. Process
Sequencing Chart

• Specific manufacturing operations

• Machine types & material quantities Route Sheets
Process Selection • Standard times (one per part)

• Process Identification
• Parts to be manufactured
• Parts to be purchased
Bill of Materials
Process Identification • Part quantities and materials Parts List
• Manufacturing processes and
Drawings
 The Make-or-Buy Decision Process
BUY
NO

Is it
Can item YES Can we YES cheaper YES Is the
be make the capital
for us to
purchased? item? available?
make?

NO NO NO YES
BUY BUY MAKE
Can we YES
make the MAKE
item

NO
 A make/buy model
• Assume a planning horizon of T periods
• In each period, the production requirement
can be satisfied by in-house production or
purchasing from a subcontractor (or from
inventories).
• An initial investment is required to have in-
house production capability.
• Inventory can be held at a cost
 A make/buy model
•  Input parameters:
– c1: unit cost of in-house production.
– c2 : unit cost of subcontracted item.
– c3 : capacity investment cost.
– K: production capacity if investment is made
– h: holding cost per item held inventory per period
• Objective is to minimize the total procurement
cost:
 A make/buy model
• Decision variables:
– xt: in house production quantity in period t.
– yt: subcontracted quantity in period t.
– It: inventory level in the beginning of period t (I1
=0).
– z: in-house production investment indicator
(z=0 if no investment is made, z=1 if
investment is made)
A simple math model
T T T
min  c1 xt   c2 yt  c3 z   hI t
t 1 t 1 t 1

s.t.
I t 1  I t  xt  yt  d t t  1,2..., T
xt  zK t  1,2..., T
It  0 t  2,3..., T
I1  0
xt  0 t  1,2..., T
z   0,1
 Bill of Materials (BOM)
• The Bill of Materials is a structured parts list
• Indicates the level of assembly at which each part
joins the product
• Lies at the heart of Material Requirements
Planning (MRP) and therefore is a key component
of Enterprise Resource Planning (ERP) software.
BOM for an air flow regulator
Assembly Order
BOM for an air flow regulator
Bill of Materials (BOM) -
example
Bill of Materials (BOM)
 A make/buy model for a given BOM
structure
A

B C(2)

• Input parameters:
– c1A, c1B , c1C : unit incremental cost of in-house
production of A, B and C.
– c2A : unit cost of subcontracted item A.
– c3 : capacity investment cost.
– KA, KB, KC : production capacity of A, B and C if
investment is made
– hA, hB, hC,: holding cost per item held inventory per
period of A, B and C
 A make/buy model for a given BOM
structure
T T T T T 1 T 1 T 1
min  c1 A x At   c1B xBt   c1C xCt   c2 A y At  c3 z   hA I At   hB I Bt   hC I Ct
t 1 t 1 t 1 t 1 t 1 t 1 t 1

s.t.
I At 1  I At  x At  y At  d t t  1..., T
I Bt 1  I Bt  xBt  x At t  1..., T
I Ct 1  I Ct  xCt  2 x At t  1..., T
I A1 , I B1 , I C1  0
x At  zK A t  1,2..., T
xBt  zK B t  1,2..., T
xCt  zK C t  1,2..., T
I At , I Bt , I Ct  0 t  2,..., T  1
x At , xBt , xCt , y At  0 t  1,2..., T
z   0,1
 Selecting required processes
• Input: Parts list indicating what is to be manufactured,
component part drawings describing each
component, quantities to be produced
– Determine all operations for each component (consider alternative
operations and raw materials)
– Identify various equipment types capable of performing the elemental
operations (manual, mechanic, automatic)
– Determine unit production times and equipment utilizations
– Evaluate alternative equipment economically
– Select processes (consider flexibility, reliability, safety, etc.)
• Output: processes, equipment, raw materials required
=> Route Sheet
 Route Sheet
OPER. OPERATION MACHINE JIGS, DEPT. STD. MACHINE
NUMBER DESCRIPTION NAME TOOLS, NO. TIME, CAPACITY MATERIAL PARTS
FIXTURES MIN. PER HR. RQD QUANT RQD QUANT

260-1 Cut to size 4 7/8” Power Saw Edge 0.45 133 Steel Pipe 8’ per
Locator 10” Dia. 10 pcs.
260-2 File Burrs Filing 0.39 158
Machine
260-3 Drill Hole Drill Press Drill 1 ½” 0.50 120
Hole
Locator
Support
260-4 Weld Gear Box Web Welding Welding 8.06 7 Gear Box Web 1
Weld Gear Box Hub Machine Equipment Gear Box Rib 1
Weld Gear Box Rib Gear Box Hub 1
Weld Screw Support No. Screw Support 1
2 No.2
Weld Augur Support Augur Support 1
260-5 Assemble Thrust Bearing Screw 1.65 36 Thrust Bearing 1
Assemble Shaft No.2 Driver Shaft No.2 1
Assemble Bronze Rubber Bronze Bearing 2
Bearing Hammer Shaft No.3 1
Assemble Shaft No.3 Oil Washer 1
Assemble Bronze Gear Box Lid 1
Bearing Screws 1
Assemble Oil Washer
Assemble Gear Box Lid
Inspection

To Motor Platform
Assembly (Op. 20-6)
 Sequencing required processes

• Assembly chart

• Operation process chart

• Precedence diagram
 TROPICAL FREEZERS CO.
ASSEMBLY CHART
CHARTED: ASSEMBLY 3 - DRIVE SHAFT AND ATTACHMENTS
PART NOS: 8,9,10,11,12,13,18,19,21,23
DATE CHARTED:____________ DRWG. NOS: 8,9,10,11,12,13,18,19,21,23
PREPARED BY: D.H.N. CHECKED BY: F.A.C.
 Operation Process Chart for air flow regulator
(Combination of route sheets and assembly chart)
Precedence Diagram for air flow regulator
 Sequencing Operations
(Principles)
• Rough work and heavy cuts – early
• Internal operations before external operations
• Critical operations with small dimensional
tolerances – late
• Physical location considerations
• Tooling and fixture considerations:
– As many operations as possible once a tool is fixed.
 Schedule Design

Quantities to Machine
SD = be produced & requirements
1. How much 1. Number of each
2. When machine type
2. Fractional
machines

1 2  n-1 n
 Quantity to be Produced

• Production quantity  Scrap estimates

• Because of scrap, target production
quantities must be larger than production
requirements
 A Binomial model
• Each item conforms to probability standards with
probability p and does not conform (and is scrapped) with
probability 1-p.
• Quality is independent for each manufacture.
• Let I be the input amount and N be the output amount. N
is a binomial random variable.

I  n
P( N  n)    p (1  p ) I  n n  0,1,2,..I
n
and
E[ N ]  Ip, Var ( N )  Ip(1  p )
 A Binomial model
•• Assume
  that the production requirement is for r items.
• Find the minimum input quantity so that the probability of
meeting the output requirements is .
 I I
I  
I  min I :  P( N  n)     p n (1  p ) I n   
*

 nr nr  n  

 • This is very easy toHow

solvedo
numerically
we find ?but does not have a
closed form formula.
A Binomial model: normal approximation

•• If
  the production requirement r is large and (1-p) is small,
the normal approximation is accurate
• the number of parts produced when the production
quantity is I, is approximately normal with:
  E[ N ]  Ip,  2  Var ( N )  Ip (1  p )
 r  Ip   r  Ip 
 P ( N  r )  P Z   1    
 
Ip(1  p)   
  Ip (1  p) 
• where Z is a standard normal random variable and F is
the standard normal cumulative distribution function.
• Let: , then is the solution of:
r  Ip
 z1
Ip(1  p)
A Binomial model: normal approximation
• Solving r  Ip for I,
 z1
Ip(1  p )
• we get:

2 pr  z 2
p (1  p )  pz (1  p )( z 2
 4 r  pz 2
)
I 
*

2 p2
• Note that the normal approximation is appropriate when
the quantities are sufficiently large. For smaller examples
better to use binomial calculation.
 Numerical Example
r p alpha z_{1-alpha}
100 0.9 0.95 -1.64485363
2 pr  z 2 p (1  p)  pz (1  p)( z 2  4r  pz 2 )
I 
*
I
100
Binomial
0
Normal
0.000429
2 p2
101
102
2.39E-05
0.000265
0.001271
0.003401
I*=117.0428
103 0.001497 0.008251
104 0.005724 0.018223
105 0.016716 0.036795
106 0.039799 0.068204
107 0.080579 0.11657
108 0.142915 0.184566
109 0.227067 0.27205
110 0.328986 0.37531
111 0.441096 0.48738
112 0.554225 0.59947
113 0.659812 0.703009
114 0.751592 0.79152
115 0.826327 0.861685
116 0.883623 0.913364
117 0.925164 0.948795
118 0.953753 0.971448 I*=118 (for %95 chance to meet target)
119 0.972495 0.984978
120 0.984233 0.99254
 Multiple Stages

I1 Operation 1 N1 Operation 2 N2

I 
P( N 2  n)    p1 p2  (1  p1 p2 ) I  n n  0,1,2,..I
n

n
and
E[ N 2 ]  Ip1 p2 , Var ( N 2 )  Ip1 p2 (1  p1 p2 )

N2 is also Binomial with probability of success p1p2

(We assume independence between the failure

rates of the two operations)
 Large scale production
• Determine production level by considering
expected number of products to produce

Ik Operation k Nk N k  I k  Pk I k
Nk
Ik 
1  Pk
I k Pk

• This is a simple approximation that is useful

when the quantities are very large and there
is time flexibility .
 Using Expectation- Multi Stages

• Pk = percent defective for machine k

• Ik = number of parts processed by machine k
• Nn = number of acceptable parts after n operations

I2 I3 In-1 In
I1 1 2  n-1 n Nn
P1 P2 Pn-1 Pn

  𝐼 = to 𝑁
Suggest a formula
𝑗 𝑛
determine
𝑛
how many to
produce∏
at each
( 1machine
− 𝑃𝑘 )
𝑘=𝑗
 CE: Find the production amount

I 10,000
 CE: Find the production amount

10,000
 10,101.01
1  0.01

12,147.04 11,418.22 10,961.49 10,413.41

 12,522.72  12,147.04  11,418.22  10,961.49
1  0.03 1  0.06 1  0.04 1  0.05
12,522.72
 12,778.28 10,000
1  0.02

10,101.01
 10,413.41
1  0.03
 CE: Find the production amount

I 10,000

2
 𝑝 =1 − [ ( 1 − 𝑝1 ) ( 1 − 𝑝 2 ) ( 1 − 𝑝3 ) ( 1 − 𝑝 4 ) ( 1 − 𝑝 5 ) ( 1 − 𝑝 6 ) ]
2
 𝑝=1− [ 0.98∗ 0 .97 ∗0.94 ∗ 0.96 ∗ 0.95∗ 0.99 ]
  22
  I=12778.3
 Small batch production
• Scrap estimates appropriate for high-volume
production
– Use of average values not appropriate for small batch
production
• How many units to produce? The following
questions may be considered:
– How much does it cost to produce a good unit?
– How much revenue is generated from a good and a bad
unit?
– What is the probability distribution for the number of
good units resulting from a batch (production lot)?
• Reject Allowance Problem: Determining the
number of additional units to produce (as a security
buffer against the scrap) when scheduling low-
volume production where faulty production (rejects)
randomly occur.
 Notation:
• Random variable representing the number of good
units produced
• Quantity of units to produce
• Probability of obtaining exactly non-faulty units if
production quantity is .
 •Cost of producing Q units, of which exactly x are
good units
• Revenue from producing Q units, of which exactly
x are good units
• Profit from producing Q units, of which exactly x
are good units

 𝑃 ( 𝑄 , 𝑥 )=𝑅 (𝑄 , 𝑥 ) − 𝐶(𝑄 , 𝑥)
 Reject Allowance Problem

• Problem:
  Determine the value of Q that
maximizes expected profit.
 Example :
• Order for exactly 20 castings – no more, no less!
• Cost per casting: 1100
• If casting not sold, recycle value of 200
• Revenue per casting: 2500
• How many castings should be scheduled for
production to maximize expected profit?
• What is the probability of losing money at this
production level?
 Example
200Q x  20 
R (Q, x)   
2500(20)  200(Q  20) 20  x  Q 

C  (Q, x )  {1100 Q 0  x  Q}

 900Q Define c x  20

P (Q, x)   
46000  900Q 20  x  Q 
 Example
19 Q
E[ P (Q)]   900Q p ( x)   (46000  900Q) p( x)
x 0 x  20

Q Q
E[ P (Q)]   46000 p( x)   900Q p( x)
x  20 x 0

Q
E[ P (Q)]   46000 p( x)  900Q
x  20
Probability distributions for the number of good
castings (x) out of Q
t02_06
Profit from producing Q casting, with exactly x being good
 Example

Optimal batch size is 28 castings. There is zero probability

of losing money at this production level.
Number of Machines Required
N = number of machines required per shift
T = standard time (minutes, hours, etc.) per unit
P = number of units to be processed per shift
E = performance level (as of % of std. time)
H = up time (per shift)
s = machine set up time
r = availability of machine, as a fraction of “up
time”

  𝑷𝑻
𝑵 =N=?
𝑬 (𝑯 − 𝒔 )𝒓
 Example
• Production volume= 20,000 per year
• Standard time = 30 minutes
• Machine utilization = 0.90
• Machine performance level = 0.95
• Daily shift = 8 hrs
• Daily set-up = 15 minutes
• Assume 52 days off per year (among 365 days in a year)
Find the fractional number of machines.
 
Stan
dard
n e ed m a ch i n .82
ed p e
er sh hours
if t

Standar
dm
machin achine hours th
e can pr
ovide p at my
er shift
 Number of Machines Required
(Individual Exercise)
Find the number of machines needed for the data given in the following table.
Part X is routed from machine A to machine B and has an annual production
volume of 100,000 units. Part Y routed from machine B to machine A and its
annual production volume is 200,000 units.
Machine A Machine B
Part X standard time (hr) 0.10 0.05
Part Y standard time (hr) 0.06 0.07
Part X scrap estimate (%) 6 3
Part Y scrap estimate (%) 5 5
Historical efficiency (performance level) (%) 95 90
Reliability factor (due to failures) (%) 80 85
Equipment availability (hr per yr) 2000 2000

X 100000
A B Y
200000
 Employee requirements
• Manual assembly operation
• Semiautomatic machines
 Number of operators needed
• Machine
  and operator engaged at all times: They are
both busy if doing something productive
• Operation j to be performed on different
products
– : number of operators for operation
– : set of products that require operation
– : production rate (pieces per shift) for product
– : average time for operation on product
– : total working hours of an operator (per shift)

   ∑ 𝑃𝑖 𝑇 𝑖𝑗
𝑖∈ 𝐼
𝐴 𝑗=
𝐶 𝑗
 Exercise
• Parts A, B and C needs painting. For these parts the
production targets per year are as follows.
• A=10000, B=20000 and C= 30000 per year
• Painting times are 5, 10 and 5 minutes, for A,B and C,
respectively
• Daily shift = 8 hrs
• Two 15-minute brakes during the shift
• Assume 52 days off per year (among 365 days in a year)
Find the fractional number of painting operators needed.
 
.83
 Semiautomatic machines

 Machines run automatically but require

manual intervention
An operator can run more than one machine

A group of operators can cooperate and run a set of machines

• Therefore, we need to create schedules for

machines and operators.
 Semiautomatic machines
f02_18

Three machines are assigned to one operator

 Example
There are three semi-automated machines.
• It takes 0.5 minute to travel from machine
to machine,
• 1 minute to load a machine and 1 minute
to unload a machine
• 6 minutes of machining time
• and 0.5 minutes to inspect and pack a part.
• What is the cycle time?
There are three semi-automated machines. It takes 0.5 minute to travel from machine to
machine, 1.0 minute to load a machine, 1 minute to unload a machine, 6 minutes of
machining time and 0.5 minutes to inspect and pack a part. What is the cycle time?

Operator loads machine 1

Time Operator Machine 1 Machine 2 Machine 3
0 L1 Loading Idle Idle
0,5 L1 Loading Idle Idle

Operator travels to machine 2

Time Operator Machine 1 Machine 2 Machine 3
0 L1 Loading Idle Idle
0,5 L1 Loading Idle Idle
1 T Machining Idle Idle

Operator loads machine 2

Time Operator Machine 1 Machine 2 Machine 3
0 L1 Loading Idle Idle
0,5 L1 Loading Idle Idle
1 T Machining Idle Idle
1,5 L2 Machining Loading Idle
2 L2 Machining Loading Idle
There are three semi-automated machines. It takes 0.5 minute to travel from machine to
machine, 1.0 minute to load a machine, 1 minute to unload a machine, 6 minutes of
machining time and 0.5 minutes to inspect and pack a part. What is the cycle time?
 Semiautomatic machines
• Multiple Activity Chart: simply a spreadsheet with
schedule (timetable) for operators and machines
• Question: How many machines should be handled by
an employee?
• Too many machines/employee
– Machine idle time
– Operator busy
• Too few machines/employee
– Machine busy
– Operator idle time
 Semiautomatic machines
Simplifying assumption: Machines are
identical
• a concurrent activity time
• b independent activity time of the
operator
• t independent activity time of the machine
 Semiautomatic machines
• a+b :time needed by operator to complete work
on a single machine in one production cycle
• a+t :time needed by a machine to complete work
in one production cycle
• t>b
• What is the number of machines assigned to an
operator for neither machine nor the operator to
be idle?

at
n 
ab
 Semiautomatic machines
Question: How many machines should I assign to
an operator?
• n' number of machines assigned to an operator
for neither to be idle
• Say n' = 3.4 (neither machine nor operator
idle)
m=3 OR m=4
operator idle time machine idle time
 Semiautomatic machines
Approach to determine best m
• Compute n'
• Check if n' is integer
– If yes, done!
– If no, pick either the largest integer less than n'
or the smallest integer greater than n'
Cost analysis
 Semiautomatic machines

•Work
  content in one production cycle:
• Operator m(a+b)
• Machine a+t

Repeating cycle time () will be the larger of

the two. Difference will be idle time.
Tc  repeating cycle time
I o  idle operator time during a repeating cycle
I i  idle time per machine i during a repeating cycle

That is,

If m  n, then Tc  a  t
If m  n, then Tc  m (a  b)
 Exercise: What is the cycle time?
• Three (identical) grinding machines are assigned to a
worker.
• Grinding takes 5 minutes at each machine
• Loading and unloading a machine takes 1 and 1.5 minutes,
respectively
• Quality check and forwarding the processed pieces to next
station takes 30 seconds
• Worker spends 30 seconds to move between machines
 
 • cost per operator-hour
• cost per machine-hour
• unit production cost if m machines are assigned to each
operator
Since each machine produces one unit during a repeating cycle
we have:
 • cost per operator-hour
• cost per machine-hour
• unit production cost if m machines are assigned to each
operator
Since each machine produces one unit during a repeating cycle
we have:
 Approach to determine best m
• Compute n'
• Check if n' is integer
– If yes, done!
– If no, pick either  n = n or
 n= n+1
• based on comparison of TC(n) and TC(n+1)
Exercise: What is the optimal number of machine
assignments to minimize the unit production cost?
Co  3
C m  10
a  2, b  1, t  5
7
TC (2)  (3  2 (10))  80.5
2
TC (3)  (3  3 (10)) 3  99
Then, m  2
 Semiautomatic machines
Actual problem (REALITY) involves uncertainty on:
• Operation time on machine
• Machine breakdowns
• Operator availability
• Operator skill level
• Contract specifications
• Production requirements (change daily or even during
shift)
• etc.