Nof S.Y. Springer Handbook of Automation
Подождите немного. Документ загружается.
Economic Rationalization of Automation Projects 40.1 General Economic Rationalization Procedure 705
Table 40.3 Example data
Project costs
Machine cost $ 80000
Tooling cost $ 13000
Software integration $ 30000
Part feeders $ 20 000
Installation cost $ 25000
Total $ 168000
Actual realizable salvage $ 12000
Table 40.4 MACRS percentages
Year Percentage
1 20.00
2 32.00
3 19.20
4 11.52
5 11.52
6 5.76
steps (Fig.40.1). To evaluate economically the automa-
tion project installation, the following data are required:
•
Capital investment of the project
•
Estimated changes in gross incomes (revenue, sales,
savings) and costs expected from the project.
To illustrate the remaining steps of the methodology,
an example will be developed. Installation cost, oper-
ation costs, and salvage for the example are as given in
Table 40.3.
Table 40.5 Costs and savings (dollars per year)
Year 1 2 3 4 5 6
Labor savings 70 000 70 000 88 000 88 000 88 000 88000
Quality savings 22000 22000 28000 28000 28000 28000
Operating costs (increase) (25000) (25 000) (19 000) (12000) (12 000) (12 000)
Table 40.6 Net cash flow
End of year K&L Total G
a
C D
b
X
0 168000 – – – −168 000
1 92 000 25000 33 600 53 640
2 92 000 25000 53 760 61 704
3 116000 19000 32256 71102
4 116000 12000 19354 70142
5 116000 12000 19354 70142
6 L = 12000 116000 12000 9677 78271
a
: These are the sums of labor and quality savings (Table 40.5)
b
: Computed with the MACRS for each year
Period Evaluation, Depreciation,
and Tax Data Requirements
Before proceeding with the economic evaluation, the
evaluation period, tax rates, and tax depreciation
method must be specified. We willconsider inthe exam-
ple an evaluation period of 6 years. We will use the US
Internal Revenue Service’s Modified Accelerated Cost
Recovery System (MACRS) for 5years (Table 40.4).
Tax rate considered is 40%. These values are not fixed
and can be changed if deemed appropriate.
Project Cost Analysis
The project cost isas given in Table 40.3(US$168000).
To continue it is necessary to determine (estimate)
the yearly changes in operational cost and cost sav-
ings (benefits). For the example these are as shown in
Table 40.5.
Economic Rationalization
Techniques used for the economic analysis of au-
tomation applications are similar to those for any
manufacturing equipment purchase. They are usually
based on net present value, rate of return or payback
(payout) methods. All of these methods require the de-
termination of the yearly net cash flows, which are
defined as
X
j
=(G −C)
j
−(G −C −D)
j
(T)−K +L
j
,
where X
j
=net cash flow in year j, G
j
=gross income
(savings, revenues) for year j, C
j
= total costs for year
j, D
j
= tax depreciation for year j, T = tax rate (as-
Part E 40.1
706 Part E Automation Management
Table 40.7 Pairwise comparison of criteria and weights
Criteria Price Weight Power Spindle Diameter Stroke
Price 1.00 0.52 0.62 0.41 0.55 0.59
Weight 1.93 1.00 2.60 1.21 1.30 2.39
Power 1.61 0.38 1.00 0.40 0.41 1.64
Spindle 2.44 0.83 2.47 1.00 0.40 0.46
Diameter 1.80 0.77 2.40 2.49 1.00 2.40
Stroke 1.69 0.42 0.61 2.17 0.42 1.00
sumed constant), K =project cost (capitalexpenditure),
and L
j
= salvage value in year j.
The net cash flows are given in Table 40.6.
Net Present Value (NPV). Once the cash flows have been
determined, the net present value (NPV) is determined
using the equation
NPV =
n
j=0
X
j
(1+k)
j
=
n
j=0
X
j
(P/F, k, j) ,
where X
j
= net cash flow for year j, n = number of
years of cash flow, k = minimum acceptable rate of re-
turn (MARR), and 1/(1+k)
j
=discount factor, usually
designated as (P/F, k, j).
With the cash flows of Table 40.6 and k = 25%, the
NPV is
NPV =−168000+53640(P/F, 25, 1)
+61704(P/F, 25, 2)+···
+78271(P/F, 25, 6) =$18431 .
The project is economically acceptable if its NPV is
positive. Also, a positive NPV indicates that the rate of
return is greater than k.
Return on Invested Capital (ROIC). The ROIC or rate of
return is the interest rate that makes the NPV =0. It is
sometimes also referred to as the internal rate of return
(IRR). Mathematically, the ROIC is defined as
0 =
n
j=0
X
j
(1+i)
j
=
n
j=0
X
j
(P/F, i, j) ,
where i = ROIC.
For this example the ROIC is determined from the
following expression
0 =−168000+53640(P/F, i, 1)
+61704(P/F, i, 2)+···+78271(P/F, i, 6) .
To solve the previous expression for i, a trial-
and-error approach is needed. Assuming 25%, the
right-hand side gives $ 18431 (NPV calculation) and
with 35% it is $ −11 719. Therefore the ROIC using
linear interpolation is approximately 31%. This ROIC
is now compared with the minimum acceptable rate
of return (MARR). In this example the MARR is that
used for calculating the NPV.IfROIC≥ MARR the
project is acceptable; otherwise it is unacceptable. Con-
sequently the NPV and the rate-of-return methods will
give the same decision regarding the economic desir-
ability of a project (investment). It is pointed out that
the definitions of cash flow and MARR are not in-
dependent. Also, the omission of debt interest in the
cash-flow equation does not necessarily imply that the
initial project cost (capital expenditure) is not being
financed by some combination of debt and equity cap-
ital. When total cash flows are used, the debt interest
is included (approximately) in the definition of MARR
as
MARR =k
e
(1−c)+k
d
(1−T)c ,
where k
e
= required return for equity capital, k
d
= re-
quired return for debt capital, T = tax rate, and c =
debt ratio of the pool of capital used for current capital
investments.
It is not uncommon in practice to adjust (increase)
k
e
and k
d
to account for project risk and uncertainties in
economic conditions.
The effects of automation on ROIC (Fig. 40.6)
are documented elsewhere in the literature [40.11].
The main effects of automation can be classified
into reduction of capital or increased profits or, more
desirably, both simultaneously. Automation may gen-
erate investment capital savings in project engineering,
procurement costs, purchase price, installation,configu-
ration, calibration or project execution. Working capital
requirements may be lowered by reducing raw material
(quantity or price), product inventories, spares parts for
equipment, reduced energy and utilities utilization or
Part E 40.1
Economic Rationalization of Automation Projects 40.1 General Economic Rationalization Procedure 707
Reduce capital Increase profit
Increased ROIC
Reduce costs Increase revenue
Increase
e price
Increase production
Reduced:
Capital investment
Product, Input,
WIP
Inventory
Warehouse
Spares
Reduced:
Energy and
utilities
Maintenance
Waste
Staff
Exceptions
Increased:
Production yield
Improved:
Products quality
Increased:
Equipment
capacity
Reduced:
Unscheduled
downtime
Scheduled
shutdowns
Fig. 40.6 Effects of automation on ROIC
increased product yields. Maintenance cost are dimin-
ished in automation solutions by reducing unscheduled
maintenance, number of routine checks, time required
for maintenance tasks, materials purchase, and num-
ber and cost of scheduled shutdown tasks. Automation
also contributes to reduce impacts (often hard to quan-
tify) due to health, safety, and environmental issues in
production systems.
Profits could increase due to automation by in-
creasing the yield of more valuable products. Reduced
work-in-process inventory and waste result in higher
revenue per unitary input to the system. Although
higher production yield will be meaningful only if the
additional products can be sold, today’s global mar-
kets will surely respond positively to added production
capacity.
Payback (Payout) Period. An alternative method used
for economic evaluation of a project is the payback
period (or payout period). The payback period is the
number of years required for incoming cash flows to
balance the cash outflows. The payback period (p)is
obtained from the expression
0 =
p
j=0
X
j
.
This isone definitionof the payback period, although an
alternative definition that employs a discounting proce-
dure is most often used in practice. Using the cash flow
giveninTable40.6, the payback equations for 2years
gives
−168000+53640+61704 =$−52656
and for 3years it is
−168000+53640+61704+71102 =$18446 .
Therefore, using linear interpolation, the payback
period is
p =2+
52656
71102
=2.74 years .
40.1.4 Additional Considerations
The following aspects must be kept in mind when ap-
plying the general procedure and related techniques
described in this section. Careful review of these issues
will ensure that the evaluation is proven correct:
•
Cash-flow equation component values are incre-
mental. They represent increases or decreases re-
sulting directly from the project (investment) under
consideration.
Part E 40.1
708 Part E Automation Management
•
The higher the NPV and rate of return, the better
(shorter) the payback period.
•
Utilization of the payback period as a primary crite-
rion is questionable since it does not consider the
cash flows generated after the payback period is
achieved.
•
When evaluating mutually exclusive alternatives,
select the highest NPV alternative. Using the high-
est rate of return is incorrect. This point is made
clear by Stevens [40.12], Blank [40.13], Thuesen
and Fabrycky [40.14].
•
When selecting a subset of projects from a larger
group of independent projects due to some con-
straint (restriction), the objective should be to
maximize the NPV of the subset of projects subject
to the constraint(s).
40.2 Alternative Approach to the Rationalization of Automation Projects
40.2.1 Issues in Strategic Justification
of Advanced Technologies
The analysis typically performed for justifying ad-
vanced technology as outlined in Sect.40.1 is financial
and short term in nature. This has caused difficulty
in adopting systems, and technology having both
strategic implications and intangible benefits usually
not captured by traditional justification approaches.
Elsewhere in the literature a number of other is-
sues that make justification and adoption of strategic
technologies difficult can be found, including high cap-
ital costs and risks, difficulty in quantifying indirect
and intangible benefits, inappropriate capital budget-
ing procedures, and technological uncertainties [40.15].
Another complicating factor in the justification of
integrated technologies is the cultural and organiza-
tional issues involved. The impact of implementing
a flexible manufacturing system crosses many orga-
nizational boundaries. The success or failure of this
implementation depends on the buy-in of all organi-
zations and individuals involved. Traditional methods
of justification often do not consider these organiza-
tional impacts and are not designed for group consensus
building.
The literature discusses many of the intangible
and nonquantifiable benefits of implementing auto-
mated systems; the most often mentioned is flexibility.
Zald [40.16] discusses four kinds of flexibility provided
by automated systems:
•
Mix flexibility: the ability to have multiple products
in the same product process at the same time
•
Volume flexibility: the ability to change the process
so that additional or less throughput is achieved
•
Multifunction flexibility: the ability to have the
same device do different tasks by changing tools on
the device
•
New product flexibility: the ability to change and
reprogram the process as the manufactured product
changes.
Other frequently mentioned benefits include improved
product quality, better customer service, improved re-
sponse time, improved product consistency, reduction
in inventories, improved safety, better employee morale,
improved management and operation of processes,
shorter cycle times and setups, and support for contin-
uous improvement and just-in-time (JIT) efforts [40.3–
6].
40.2.2 Analytical Hierarchy Process (AHP)
Rarely do automation systems comprise out-of-the-box
solutions. In fact, most automated systems nowadays
comprise a collection of equipment properly integrated
into an effective solution. This integration process
makes evaluation of alternative solutions more com-
plex, due to the many combinations possible for the
configuration of all available components. To assist
the equipment selection process AHP has been im-
plemented in the form of decision support systems
(DSSs) [40.17]. AHP was developed by Saaty [40.18]
as a way to convey the relative importance of a set of
activities in a quantitative and qualitative multicriteria
decision problem. The AHP method is based on three
principles: the structure of the model [40.19], compar-
ative judgment of alternatives and criteria [40.20], and
synthesis of priorities. Despite the wide utilization of
AHP, selectionof casting process [40.21], improvement
of human performance in decision-making [40.19], and
improvement of quality-based investments [40.22], the
method has shortcomings related to its inability to han-
dle decision-maker’s uncertainty and imprecision in
determining values for the pairwise comparison process
involved. Another difficulty with AHP liesinthefact
Part E 40.2
Economic Rationalization of Automation Projects 40.2 Alternative Approach to the Rationalization of Automation Projects 709
Forming decision-making team
Determining alternative equipment
Structuring decision hierarchy
Determining the criteria to be used
in evaluation
Approve
decision
hierarchy
?
Stage 1: Data gathering
Stage 2: AHP calculations
No
Assigning criteria weights via AHP
Partial ranking via PROMETHEE I
Structuring decision hierarchy
Approve
decision
weights
?
No
Step 1:
Step 2:
Step 3:
Step 4:
Step 5:
Step 6:
Step 7:
Determining the preference functions
and parameters for the criteria
Approve
preference
functions
?
No
Step 8:
Step 9:
Step 10:
Complete ranking via PROMETHEE IIStep 11:
Determining GAIA planeStep 12:
Determining the best equipmentStep 13:
Stage 3: PROMETHEE calculations
Stage 4: Decision making
Fig. 40.7 Steps of AHP-PROMETHEE method
Selection of
the best equipment
Machine 3Machine 2Machine 1 Machine 4 Machine 5
Price Weight Power Spindle Diameter Stroke
Fig. 40.8 Decision structure
Part E 40.2
710 Part E Automation Management
Table 40.8 AHP results
Criteria Weights (ω) λ
max
, CI, RI CR
Price 0.090 λ
max
=6.201 0.032
Weight 0.244
Power 0.113 CI =0.040
Spindle 0.266
Diameter 0.186 RI =1.24
Stroke 0.101
CR: Consistency ratio
RI: Random index
CI: Consistency index
that not every decision-making problem may be cast
into a hierarchical structure.
Next, a proposed method implementing AHP is
reviewed using a numerical application for computer
numerical control (CNC) machine selection.
The AHP-PROMETHEE Method
The preference ranking organization method for en-
richment evaluation (PROMETHEE) is a multicri-
teria decision-making method developed by Brans
et al. [40.23, 24]. Implementation of PROMETHEE
requires two types of information: (1) relative im-
portance (weights of the criteria considered), and
(2) decision-maker’s preference function (for compar-
ing the contribution of alternatives in terms of each
separate criterion). Weights coefficients are calculated
in this case using AHP. Figure 40.7 presents the various
steps of the AHP-PROMETHEE integration method.
The AHP-PROMETHEE method is applied to
a manufacturing company wanting to purchase a num-
ber of milling machines in order to reduce work-in-
process inventory and replace old equipment [40.17].
A decision-making team was devised and its first task
was to determine the five milling machines candi-
dates for the purchasing and six evaluation criteria:
Table 40.9 Evaluation matrix for the milling machine case
Criteria Price Weight Power Spindle Diameter Stroke
Unit US$ kg W rpm mm mm
Max/min Min Min Max Max Max Max
Weight 0.090 0.244 0.113 0.266 0.186 0.101
Machine 1 936 4.8 1300 24000 12.7 58
Machine 2 1265 6.0 2000 21000 12.7 65
Machine 3 680 3.5 900 24000 8.0 50
Machine 4 650 5.2 1600 22 000 12.0 62
Machine 5 580 3.5 1050 25 000 12.0 62
Table 40.10 Preference functions
Criteria PF Thresholds
q p s
Price Level 600 800 –
Weight Gaussian – – 4
Power Level 800 1200 –
Spindle Level 20000 23000 –
Diameter Gaussian – – 6
Stroke V-sh ape – 50 –
Table 40.11 PROMETHEE flows
Alternatives Φ
+
Φ
−
Φ
Machine 1 0.0199 0.0139 0.0061
Machine 2 0.0553 0.0480 0.0073
Machine 3 0.0192 0.0810 −0.0618
Machine 4 0.0298 0.0130 0.0168
Machine 5 0.0478 0.0163 0.0315
price, weight, power, spindle, diameter, and stroke. The
decision structure is depicted in Fig. 40.8. The next
step is for decision team experts to assign weights on
a pairwise basis to decision criteria, as presented in Ta-
ble 40.7. Results from AHP calculations are shown in
Table 40.8 and show that the top three criteria for the
case are spindle, weight, and diameter. The consistency
ratio of the pairwise comparison matrix is 0.032 < 0.1,
which indicates weights consistency and validity.
Following the application of AHP, PROMETHEE
steps are carried out. The first step comprises the eval-
uation of five alternative milling machines according to
the evaluation criteria previously defined. The resulting
evaluation matrix is shown in Table 40.9.
Next, a preference function (PF) and related thresh-
olds are defined by the decision-making team for each
criterion. PF and thresholds consider features of the
milling machines and the company’s purchasing pol-
Part E 40.2
Economic Rationalization of Automation Projects 40.3 Future Challenges and Emerging Trends in Automation Rationalization 711
1
Machine 5
Φ+ 0.05
Φ– 0.02
3
Machine 2
Φ+ 0.06
Φ– 0.05
2
Machine 4
Φ+ 0.03
Φ– 0.01
4
Machine 1
Φ+ 0.02
Φ– 0.01
5
Machine 3
Φ+ 0.02
Φ– 0.08
Fig. 40.9 PROMETHEE I partial ranking
icy. Table 40.10 shows preference functions and their
thresholds.
The partial ranking of alternatives is determined
according to PROMETHEE I, based on the positive
and negative flows shown in Table 40.11. The result-
ing partial ranking is shown in Fig. 40.9 and reveals that
machine 5, machine 2, machine 4, and machine 1 are
preferred over machine 3, and machine 4 is preferred
over machine 1. The partial ranking also shows that ma-
chine 5, machine 4, and machine 2 are not comparable,
as well as machine 5 and machine 1, and machine 2 and
machine 1.
PROMETHEE II uses the net flow in Table 40.11 to
compute a complete ranking and identify the best alter-
native. According to the complete ranking, machine 5
is selected as the best alternative, while the other ma-
chines are ranked accordingly as machine 4, machine 2,
machine 1, then machine 3 (Fig.40.10).
2
Machine 4
Φ 0.02
1
Machine 5
Φ 0.03
3
Machine 2
Φ 0.01
5
Machine 3
Φ –0.06
4
Machine 1
Φ 0.01
Fig. 40.10 PROMETHEE II complete ranking
Criterion 1
Criterion 5
Criterion 6
Criterion 2
Criterion 4
Criterion 3
Machine 2
Machine 3
Machine 5
Machine 1
Machine 4
π
Fig. 40.11 GAIA decision plane
The geometrical analytic for interactive aid (GAIA)
plane [40.25] representing the decision (Fig.40.11)
shows that: price has great differentiation power,
criteria 1 (price) and 3 (power) are conflicting, ma-
chine 2 is very good in terms of criterion 3 (power),
and machine 3 is very good in terms of crite-
ria 2 and 4. The vector π (decision axis) represents
the compromise solution (selection must be in this
direction).
40.3 Future Challenges and Emerging Trends
in Automation Rationalization
40.3.1 Adjustment of Minimum Acceptable
Rate of Return in Proportion
to Perceived Risk
Since capital investments involve particular levels of
risk, it is common practice for management to increase
the MARR for automation projects involving higher
risk. Assigning a higher MARR forces the proposed
projects to generate greater return on their capital in-
vestment. A similar strategy is proposed to recognize
the fact that capital equipment that can be reused for
additional projects is less likely to experience a decline
in its value due to unforeseen reductions in a given
product’s demand and market life.
Traditionally, more stringent MARR requirements
are applied to the entire capital investment. It is pro-
Part E 40.3
712 Part E Automation Management
posed that only the portion of the capital investment
that is at risk due to sudden changes in market de-
mand or operating conditions should be forced to meet
these more demanding rate of return. This approach
to assigning risk will explicitly recognize and promote
the development and reuse of reconfigurable automa-
tion by compensating for its higher initial development
and implementation costs through lower rate-of-return
requirements.
40.3.2 Depreciation
and Salvage Value Profiles
Mechanisms for capital depreciation and estimated
project salvage values also have a significant effect on
the financial justification of automation. Two mecha-
nisms of depreciation must be considered: tax and book
depreciation. Tax depreciation methods provide a sys-
tematic mechanism to acknowledge the reduction of
capital asset value over time. Since depreciation is tax-
deductible, it is generally in the best interests of the
company to depreciate the asset as quickly as possi-
ble. Allowable depreciation schedules are determined
by the country tax code. Tax depreciation can become
a factor when comparing product-specific automation
for reconfigurable systems when the projected prod-
uct life is less than the legislated tax depreciation life.
Under these circumstances the product-specific systems
can be sold on the open market and the remaining tax
depreciation would be forfeited. A second alternative is
that the system would be decommissioned and stored
by the company until it is fully depreciated and then
sold for salvage. Instead, it is unlikely that the life of
a reconfigurablesystem wouldbe shorter than the estab-
lished tax depreciation period due to its redeployment
for a new application project. Tax depreciation terms are
determined by tax and accounting conventions rather
than the expected service life of the asset. Book depre-
ciation schedules are therefore developed to predict the
realizable salvage value at the end of an asset’s useful
life. Two approaches for determining realizable salvage
value can be utilized: the internal asset value in the
organization (when it is used in an additional project)
and the asset value when sold for salvage. Automation
equipment profitably redeployed within the same com-
pany is clearly more valuable to that organization than
to an equipment reseller.
Product-specific automation, when sold off for
scrap, has value for potential buyers only due to
their interest in key system components. Product- and
process-specific tooling will represent very little value
for equipment resellers and, most likely, all the custom
engineering and software development required to field
the system will be lost. On the other hand, if the sys-
tem has been developed based on modular components,
well understood by the user and other manufacturing
organizations, it may represent more value than a com-
pletely new set of system components. A redeployed
system may result in lower time and cost to provide
useful automation resources. Investments made in the
development of process technology may be of value
in subsequent projects [40.26]. Application software, if
developed in a modular fashion, also has the potential
for reutilization.
40.4 Conclusions
Although computation of economic performance indi-
cators for automation projects is often straightforward,
rationalization of automation technology is fraught
with difficulty and many opportunities for long-term
improvements are lost because purely economic eval-
uation apparently showed no direct economic benefit.
Modern methods, taking into account risks involved
in technology implementation or comparison of com-
plex projects, are emerging to avoid such high-impact
mistakes. In this chapter, in addition to providing
the traditional economic rationalization methodology,
strategic considerations are included plus a discussion
on current trends of automation systems towards agility
and reconfigurability.
Part E 40.4
Economic Rationalization of Automation Projects References 713
References
40.1 S. Brown, B. Squire, K. Blackmon: The contribution
of manufacturing strategy involvement and align-
ment to world-class manufacturing performance,
Int. J. Oper. Prod. Manag. 27(3-4), 282–302 (2007)
40.2 A.M.A. Al-Ahmari: Evaluation of CIM technologies
in Saudi industries using AHP, Int. J. Adv. Manuf.
Technol. 34(7-8), 736–747 (2007)
40.3 K. Feldmann, S. Slama: Highly flexible assembly –
scope and justification, CIRP Ann. Manuf. Technol.
50(2), 489–498 (2001)
40.4 D. Dhavale: Justifying manufacturing cells, Manuf.
Eng. 115(6), 31–37 (1995)
40.5 J.F. Kao, J.L. Sanders: Analysis of operating policies
for manufacturing cells, Int. J. Prod. Res. 33(8),
2223–2239 (1995)
40.6 R. Quaile: What does automation cost – calculat-
ing total life-cycle costs of automated production
equipment such as automotive component man-
ufacturing systems isn’t straightforward, Manuf.
Eng. 138(5), 175 (2007)
40.7 D. Vazquez-Bustelo, L. Avella, E. Fernandez: Agility
drivers, enablers and outcomes – empirical test of
an integrated agile manufacturing model, Int. J.
Oper. Prod. Manag. 27(12), 1303–1332 (2007)
40.8 J.J. Mills, G.T. Stevens, B. Huff, A. Presley: Justifica-
tion of robotics systems. In: Handbook of Industrial
Robotics, ed. by S.Y. Nof (Wiley, New York 1999)
pp. 675–694
40.9 G. Boothroyd, C. Poli, L.E. Murch: Automatic As-
sembly (Marcel Dekker, New York 1982)
40.10 S.Y. Nof, W.E. Wilhelm, H.-J. Warnecke: Industrial
Assembly (Chapman Hall, London 1997)
40.11 D.C. White: Calculating ROI for automation projects,
Emerson Process Manag. (2007), available at
www.EmersonProcess.com/solutions/
Advanced Automation (Last access date: March 20,
2009)
40.12 G.T. Stevens Jr.: The Economic Analysis of Capital
Expenditures for Managers and Engineers (Ginn,
Needham Heights 1993)
40.13 L.T. Blank, A.J. Tarquin: Engineering Economy,6th
edn. (McGraw-Hill, New York 2004)
40.14 G.J. Thuesen, W.J. Fabrycky: Engineering Economy,
9th edn. (Prentice Hall, Englewood Cliffs 2000)
40.15 O. Kuzgunkaya, H.A. ElMaraghy: Economic and
strategic perspectives on investing in RMS and FMS,
Int. J. Flex. Manuf. Syst. 19(3), 217–246 (2007)
40.16 R. Zald: Using flexibility to justify robotics automa-
tion costs, Ind. Manag. 36(6), 8–9 (1994)
40.17 M. Da
ˇ
gdeviren: Decision making in equipment se-
lection: an integrated approach with AHP and
PROMETHEE, J. Intell. Manuf. 19, 397–406 (2008)
40.18
T.L. Saaty: The Analytic Hierarchy Process (McGraw-
Hill, New York 1980)
40.19 E. Albayrak, Y.C. Erensal: Using analytic hierarchy
process (AHP) to improve human performance: an
application of multiple criteria decision making
problem, J. Intell. Manuf. 15, 491–503 (2004)
40.20 J.J. Wang, D.L. Yang: Using hybrid multi-criteria
decision aid method for information systems out-
sourcing, Comput. Oper. Res. 34, 3691–3700 (2007)
40.21 M.K. Tiwari, R. Banerjee: A decision support system
for the selection of a casting process using analytic
hierarchy process, Prod. Plan. Control 12,689–694
(2001)
40.22 Z. Güngör, F. Arikan: Using fuzzy decision mak-
ing system to improve quality-based investment,
J. Intell. Manuf. 18, 197–207 (2007)
40.23 J.P. Brans, P.H. Vincke: A preference ranking
organization method, Manag. Sci. 31,647–656
(1985)
40.24 J.P. Brans, P.H. Vincke, B. Mareschall: How to select
and how to rank projects: the PROMETHEE method,
Eur. J. Oper. Res. 14, 228–238 (1986)
40.25 A. Albadvi, S.K. Chaharsooghi, A. Esfahanipour:
Decision making in stock trading: an applica-
tion of PROMETHEE, Eur. J. Oper. Res. 177,673–683
(2007)
40.26 S.L. Jämsä-Jounela: Future trends in process au-
tomation, Annu. Rev. Control 31(2), 211–220 (2007)
Part E 40
“This page left intentionally blank.”