See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/226300115

Cost Estimation and Conceptual Process Planning

Chapter · January 2007

DOI: 10.1007/978-0-387-49864-5_28

CITATIONS READS
6 5,048

3 authors:

Patrick Martin Jean-Yves Dantan

Ecole Nationale Supérieure d'Arts et Métiers Metz Ecole Nationale Supérieure d'Arts et Métiers
197 PUBLICATIONS   1,201 CITATIONS    122 PUBLICATIONS   1,313 CITATIONS   

SEE PROFILE SEE PROFILE

A. Siadat
Ecole Nationale Supérieure d'Arts et Métiers
164 PUBLICATIONS   1,176 CITATIONS   

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Inspection planning in multistage manufacturing systems View project

Master Thesis View project

All content following this page was uploaded by Patrick Martin on 07 January 2014.

The user has requested enhancement of the downloaded file.

 COST ESTIMATION AND CONCEPTUAL
PROCESS PLANNING

P. Martin, J.-Y. Dantan, A. Siadat, X. Houin, Q. Daniel

Laboratoire de Génie Industriel et de Production Mécanique
Ecole Nationale Supérieure d’Arts et Métiers (ENSAM)
[email protected]

Engineering cost estimation is now compulsory from the very first stages of
design. The later a cost issue will be detected, the more it will cost. This paper
tends to show how to define which information is needed to allow this
estimation to be done. A taxonomy that builds a structure in this information
has been created. Using this taxonomy, we designed a method to calculate
manufacturing costs thanks to an expert system. The cost is determined by
considering Cost Entities using pertinent inductors and parametric calculation
methods.

1. INTRODUCTION
Cost considerations are nowadays critical in the engineering field. We know how
tiny the liberty of action and decision for companies is. We also know how to
evaluate the costs engagement
during a project. During the conceptual design stage, the decisions we make
engage a lot of money for the following stages of the future. This is the stage where
most of the choices that imply a cost are made. However, it is also the stage where
changing the decisions is the cheapest. When a company has to bring modifications
on the design while in the industrialisation stage, it is very costly because all the
design process has to be done again.
Therefore it is necessary to bring to designers as much information as possible so
that they can take the accurate decisions. In this context, letting designers to have
in the very first stages of design a tool to evaluate the costs can be very attractive. It
can prevent an inaccurate design that would go to the industrialisation department,
which would send it back to designers because it is too expensive to manufacture.
We will then try to define a tool that could handle these tasks.

2. CONCEPTUAL PROCESS PLANNING/CONCEPTUAL

DESIGN
This chapter is based on research by Shaw C. Feng research [1]. It deals with the
means to determine a Conceptual process design from the conceptual design.
2 Digital Enterprise Technology

To evaluate the manufacturing costs, it seems obvious that we need to know

exactly what processes will be used to manufacture the part. However, in the
conceptual design stage, we have no idea of what the processes will be. The aim of
this chapter is to show how we can define a conceptual process design in the very
first stages of a project.

2.1 Link between Conceptual Design and Conceptual Process Planning

Conceptual design refers to a stage in the design procedure. This stage is the one
where designers define rough part characteristics based on the part requirements.
Information such as the material, the general part shape and main geometrical data is
provided during this stage. However, data are still rough and do not allow to define
extensively the manufacturing processes.
Conceptual process design is a first estimation of what the process plan could be.
Manufacturing operations are not very detailed but it allows engineers to have an
idea of processes and attached resources. These data make a first cost evaluation
possible. These results will allow designers to adapt their decisions regarding the
costs, which is something they can not do for the time being.
Now the problem remains how to define the conceptual process plan from the
conceptual design. Figure 1 shows how the two stages can be performed
simultaneously.
After this stage, the project will continue with a detailed design stage. The
interest in the method is to start the detailed design stage with a validated conceptual
design. This will prevent the detailed design stage to be done several times.
Shaw C. Feng proposed a method to generate the Conceptual Process Planning
from the Conceptual Design data. Three main steps can be found in this method. All
of them are critical in the cost evaluation process.

2.2 Manufacturing processes selection [2]

This activity can be divided in four activities. Its aim is to select processes that are
compatible with the data we have obtained from the Conceptual Design stage. The
different activities are the following:
- Determine compatible processes based on materials
- Determine compatible processes based on quantity
- Determine compatible processes based on product shape
- Determine compatible processes based on tolerances.

For this selection process, we will consider as potential processes the total range
of processes. Then the material will prevent some of these processes. So the
outcome will be a narrowed range of compatible processes. For example, we know
we will not propose to forge a plastic part. Thanks to these four steps in a row, we
will come out with a reduced range of processes that will be proposed for the
Conceptual Process Plan.
Once these compatible processes are selected, we could rank them in terms of
relative cost and finally let the user choose those he will consider for the upcoming
stages.
Digital Enterprise Technology 3

Conceptual Design Conceptual

Process Planning

Form/Structure Process
Requirement & property Process Selection
•Fabrication
Functional •Assembly
Design •Inspection
Function Resource
Behavioral Selection
Specification
Estimated Equipment/Skill
Behavior Time & Cost
Time & Cost
Embodiment
Estimation
Design
Form/Structure Time & Cost

Detailed Detailed Process

Design Planning
•Geometry
•Operation sequences
•Topology
•Process parameters
•Tolerances
•Setup/Fixture
•Dimensions
•Accurate time& cost
•Surface Conditions
•Manufacturing resource
•Materials

Figure 1 - Interaction between Conceptual Design and

Conceptual Process Planning [1]

Processes are now considered known. If we want to evaluate the induced cost,
we also need to determine what will be the used resources: type of machines,
toolings, tool, fixtures, labour skills.

2.3 Manufacturing cost estimation

Once the manufacturing processes and the manufacturing resources have been
selected, we have the information to evaluate the cost. We have to keep in mind that
this calculated cost will be a rough evaluation. The purpose is not to determine the
actual manufacturing cost but to assess manufacturability and cost in the early
design stage. For this estimation stage, we propose another method than Shaw C.
Feng’s one. This method will be developed in part 3.

3. COST ESTIMATION
3.1 Cost evaluation methods

There are several methods that exist for calculating costs. Table 1 presents the
main techniques. Accordingly with this table, in the conceptual design stage, the
best methods to apply are analogical or parametric. Indeed, we do not have much
information on the product yet and an analytical method would be inappropriate.
However, these methods might give too rough estimation to allow the designer
checking the validity of the product he has started to design. These methods may not
help the designer very much in his tasks.
Therefore, we will adapt an analytical method to the conceptual design stage. As
we have already shown above, we can determine what processes and what resources
4 Digital Enterprise Technology

can be used. This means that from the Conceptual Design stage, we have deduced
some information usually provided in the detailed design stage. Yet, the information
is still not accurate enough to allow an usual analytical method.

Table 1: Main cost estimation techniques [3]

Best applicable
Description Accuracy
for
Intuition Evaluation of cost regarding personal Preliminary from -30%
method knowledge and intuition. stage to 50%

Preliminary from -30%

Comparison Evaluation using similar parts
stage to 50%
Case-based evaluation, definition of
Analogical Conceptual from -14%
main parameters for comparison with
method design to 30%
previous cases
One or several parameters are chosen
Parametric Conceptual from -14%
to be critical. They are used along with
method design to 30%
coefficients to evaluate the cost
Direct and indirect costs are
Analytical considered. Each cost is calculated from -5%
Detailed design
method and then they are all summed to get to 15%
the product cost

3.2 Adapted cost entity method

The cost entity method was created by H’Mida [4] in the context of a PhD at
ENSAM Metz. It is an evolution of the well-known ABC method. These methods
are analytical methods.
The cost entity method separates the company in several activities. To each
activity corresponds a cost entity. Figure 2 represents how this cost entity is defined.

Input object Resources

Cost
Cost entity
driver

Output
activity
Figure 2 - Cost entity model [4]

Each cost entity is calculated by using equation 1. We can note that there is
homogeneity between all the activities which is one of the advantages of the method.
All activities can be considered identically in terms of cost contribution form.
Cost entity = d x Σ ( ar x IRR)
d: Unique driver chosen for the Cost Entity
αr: Resource R consumption coefficient
IRr: Resource R imputation rate
Digital Enterprise Technology 5

To adapt this method to conceptual design, we will simplify the formula by

selecting one driver and one linked coefficient for the whole cost entity. The
coefficient can be a property of a machine or a material for example. As these
coefficients may be difficult to evaluate, the company may use old cost estimations
to determine them. Then the used formula reduces to the following equation :
Cost entity ( ) = Driver ( type) x Imputation rate ( / type) + fixed cost ( )

Roughly speaking, this method results in using the structure of a cost entity
model with a parametric method. However here the parametric method is not
applied to the part but to the cost entity.

3.3 Considered cost entities

To represent all the costs we will consider in our evaluation, we use a taxonomy.
This representation details what are the components of the Part Cost Entity and what
they are composed of. This drawing is presented in figure 3.
This scheme shows that we consider both direct costs and indirect costs with the
same importance. Thanks to this cost entity, we will be able to evaluate the overall
cost of the product while we are just in the conceptual design stage.
The table 2 shows examples of cost entities expressions. Yet the table is not
exhaustive and some additional research has to be done to determine every cost
entity expression.

Figure 3 - Cost Entity taxonomy

6 Digital Enterprise Technology

Table 2: Cost entities examples

driver description depends on
ρ material density
part volume material
(kg/m^3)
Rough Vb rough part volume
part
material cost (m^3)
Cunit material unit cost
formula material
per kilo ( /kg)
Cmaterial = Ir x
Ir : filling index (<1) part
Vb x ρ x Cunit
driver description depends on
Entity complexity CM : machine cost rate
machine
index ( / h)
Launching A : configuration time
per machine machine
coefficient (h / Ie)
cost Ie : entity complexity
formula part
index
Csetup=Ie x A x depends on topology,
CM tolerances….
driver description depends on
Quality per tolerance IT : tolerance (mm) part
tolerance
cost formula b : coefficient ( .mm) machine
Ctol=a+b/IT a : coefficient ( ) machine
driver description depends on
Administrativ production size Q : production size (qty) part
e cost per administrative
formula k : coefficient ( /qty)
batch operation
Cadmin = Q x k + F : fixed cost coefficient administrative
F ( ) operation
driver description depends on
Transfer amount Q : production size (qty) part
Material S : part projected surface
handling cost formula part
(m^2)
per material Chandling={(QxS) Strans : transportation
handling facility
/Strans]xB surface (of truck) (m^2)
B : coefficient ( /transfer) handling facility

4. SYSTEM IMPLEMENTATION

If we can go through the two stages we have shown so far, we would be able to
evaluate with a good accuracy the cost part in the early design stage. These two
stages are the conceptual process planning and the cost entity model calculation.
However, these steps are difficult steps to go through. In this chapter, we will
show how an expert system can help with the process implementation.
Digital Enterprise Technology 7

4.1 Ontology and taxonomy

Before implementing the estimation process with an expert system, we have to

structure the information. This can be made thanks to an ontology and a taxonomy.
The ontology is used to represent the vocabulary linked to part manufacturing
field. It contains object classes such as Tools or people or concepts such as Costs. It
also contains the different attributes that can be added to these classes (description,
size, velocity …) and the relations between objects (a Machining Operation can be
linked to a Machine by the relation “is made by”). To structure it, we will use a
taxonomy. This graphic representation exposes clearly the hierarchy and class
instances.

4.2 Application to conceptual process planning

The data we will deal with in conceptual process planning can be classified in three
main categories. First will be data related to the entity (general information on the
product), then data related to the processes and finally data related to the resources.
A detailed version of the “Operation” branch is presented on figure 4. We will
not develop here the other branches because it will not be of a great importance in
the presentation. This hierarchy is taken from Marty and Linares work upon
manufacturing processes [5].
Thus, while selecting the processes thanks to the previously viewed method, we
will look for processes among these. We consider it is not necessary to continue
with a finest description of processes (“Drilling” can be “Short drilling” or “Deep
drilling” for example) because we will not have the information to satisfy such a
level of detail. Besides, the described precision will be accurate enough to have a
general view upon manufacturing costs.

Figure 4 - Operation branch [5]

 8 Digital Enterprise Technology

4.3 Expert system development

To implement the system, we chose to use the “C Language Integrated Production

System” (CLIPS) [6]. This system was created by the NASA. CLIPS provides a
complete environment for the construction of rule and/or object based expert
systems.
So by entering all these rules in the expert system, we will be able to select what
are the compatible processes. Afterwards the user will be asked for the one he wants
to use and we will then know the process and will be able to evaluate the cost.

5. CONCLUSION
We presented in this paper how we can evaluate parts costs from the early design
stage. This activity would be very attractive for manufacturers and may make
companies save time and money.
This evaluation process implies a conceptual process planning. We have
described how we can link conceptual design to conceptual process planning.
However, the implemented program deals only with the processes selection. The
program is not achieved as resources selection and cost calculation are lacking.
Nevertheless the theory has been shown.
We have also described the method we may apply to calculate the costs. This
method would allow considering the global package of costs that could be linked to
the part manufacturing. It means not only direct and manufacturing costs are taken
into account but also indirect costs such as handling. However, this method needs to
have a high number of coefficients and a good knowledge of costs is necessary to
allow these coefficients calculation.

6. REFERENCES

1. Boothroyd G., Dewhurst P., Knight W., Product design for manufacture and assembly, Textbook,
Edition Marcel Dekker Inc, University of Rhode Island ,1994
2. C. Marty, M. Linares, procédés de fabrication, Vol. 3, Industrialisation des produits mécaniques,
édition Paris hermés, 1999
3. Feng Shaw C., Conceptual Process Planning, a Definition anf Functionnal Decomposition,
Manufacturing Science in Engineering, V10 in the proceedings of the international mechanical
engineering congress and exposition , 2000
4. Mida F., Martin P., Vernadat F., Cost estimation in mechanical production : the cost entity approach
applied to integrated product engineering, International journal of production economics (in press)
5. http://www.ghg.net/clips/CLIPS.html
6. R.C. Creese & L.T. Moore "Cost Modelling for concurrent engineering", Cost engineering, Vol. 32,
1990

View publication stats