Apem2 4 - 149 162 PDF
Apem2 4 - 149 162 PDF
Apem2 4 - 149 162 PDF
Welding feature: ES
2
ij
Preparing feature: EP
2
ij
Welding feature: ES
k
ij
Preparing feature: EP
k
ij
Masmoudi, Hachicha, Bouaziz: Development Of A Welding Cost Estimation Model Based On
152
3.2. Feature model
Preparing features
The preparing feature includes the machining of borders in V, X etc and also the
positioning of the parts to weld one according to the other (gap, alignment, pointing, etc). It is
principally defined by the geometric shape, which defines the space to be filled with the filling
metal. Table I presents examples of preparing features in V, in straight borders and in X.
Among these preparing features, there are some, which are technically and economically
recommended, we mention here: recommended preparing features.
Table I: An example of a preparing feature parameter.
Diagram Features parameters
Preparing feature
"End to End in V"
Thickness e
Basic metal steel
Preparation form En V
Chamfer opening angle
Clearance apace g
Heel or height of the flat t
Preparing feature
"End to End
with vertical borders"
Thickness e
Basic metal steel
Preparation form En l
Clearance apace g
Preparing feature
"End to End in X"
Thickness e
Basic metal steel
Preparation form En X
Chamfer opening angle
Clearance apace g
Heel or height of the flat t
Welding feature
According to the welding process and techniques, the weld joint is obtained in one or
several processes executed by the operator. The welding feature is principally defined by the
parameter presented in Table II.
Table II: Parametring of a welding feature
Technical parameters
Basic metal Example : steel
Thickness e (mm)
Assembling type Example : End to End
Welding position Example : Flat
Welding process Example : SMAW
Number of re-starting z
Welding intensity I (A)
Type of current Example : continued
Electrode diameter (mm)
Electrode output
Protection gas density D
Geometric parameters
Weld section SS
Height of weld h
Masmoudi, Hachicha, Bouaziz: Development Of A Welding Cost Estimation Model Based On
153
4. TIME AND COST CALCULATION
4.1. Analytic method calculation of the welding time
To estimate the feature welding time for an assemblage, we propose an approach based on
the evaluation of the welding section from a geometric modelling.
Geometric modelling
Fig. 3 presents the approach to calculate the welding section based on the geometric
parameters of the corresponding feature preparing.
Figure 3: The geometric modelling approach of a welding section.
The geometric modelling related to each preparing feature must be saved in a technical
data base. To explain the approach we are going to treat the example of preparing feature
end to end in V. Fig. 4 presents an example of a geometric modelling in order to determine
the welding section.
Figure 4: Geometric modelling of a welding feature "End to End" in V.
The preparing feature section marked "SP" is represented by two times the surface T.
The theoretical welding section is given by the fallowing equation:
T R B SST 2 + + = (1)
To determine "SST", it is necessary to calculate:
- the two triangles section "T" which is expressed by:
t) (e
g) (l
T
=
2
2 (2)
where
)
tg ( t) (e g l
2
2 + = (3)
Preparing feature EP
Welding section
theoretical: SST
) , , h,t,form g f(e, SST =
Welding feature
ES
Welding section
corrected: SS
SST SS =
) , , form t f(e, SP =
g
l
g
h
t
Part i
R
T T
B
Part j
e
Masmoudi, Hachicha, Bouaziz: Development Of A Welding Cost Estimation Model Based On
154
- R which is the triangle surface with a length "g" and with "e",
e g R = (4)
- B is estimated at 75 % of the rectangle section of a length "l" and a width "h". So:
h l B =
4
3
(5)
According to (1), we can write that:
h l )
tg( t) (e e g SST + + =
4
3
2
2
(6)
By replacing the expression of "l" in the equation (6), we get :
h g )
tg( h) t (e t) (e e g SST + + + =
4
3
2 2
3
(7)
Generally, the welder cant respect exactly the limits of the preparing section while
executing the welding operation. For this reason, we suppose that the section SST must be
corrected by a coefficient "". The relation between the theoretical SST and the actual section
is written:
SST SS = (8)
"" is practically near the unity in the case of an automatic application of the welding
operation.
Welding time estimation
The principle consists in establishing a simple cost estimating system which closely
associates the technical parameters (welding features), and the geometric parameters. To
determine a feature welding time, it is necessary to determine the geometric parameters so
as to calculate the volume of the filling metal in the welding area. When we know the basic
metal density, we can determine the weight of the weld joint and then the welding time see
the organigram present in Fig. 5.
Figure 5: Welding time calculating methodology.
Filling metal weight
Welding joint
Filling metal mass rate: D
a
A feature arc time: t
arc
Length of weld: L
Preparing feature
- Welding process:,
- Welding intensity: I
Welding feature
Feature welding time: t
s
- Position coefficient: p
- Filling metal density: d
Technical
Data Base
Geometric modelling: SS
Masmoudi, Hachicha, Bouaziz: Development Of A Welding Cost Estimation Model Based On
155
The use of time estimating analytic method can be perfectly considered. The estimating
approach of the welding time is organized around the following four stages:
Stage 1: welding volume evaluation. It is the produce of the welding section by the
welding length:
ij s
.L SS V = (9)
Where L
ij
: welding length
SS : welding section
Stage 2: Amount of Welding Wire
.d V
m
s
a
= (10)
Where d : the filling metal density
: the process efficiency
: the electrode output
Stage 3: The electrode time "t
arc
" is the electrode and protecting gas consuming time. It
is expressed by:
a
a
arc
D
m
t = (11)
Stage 4: the welding operation executing time is the arc time corrected thanks to the
operator efficiency and to the coefficient of the positioning difficulty:
arc S
t
p t = (12)
Where : the process depending on the operator efficiency;
p : this coefficient depends on the complexity of the weld positioning and of the
assemblage type.
Basing on the formulas (9), (10), (11) and (12), for one feature, we can write:
ij
a
S
L SS .
D
d
p
t
(13)
4.2. Cost estimation
Definition of the cost feature approach
We adopt the cost feature definition proposed by [5] and [6]: "A feature cost is a
grouping of costs associated to resources consumed by one activity. The fundamental
condition cares about the homogeneity of the resources consumed by the feature cost which
allows associating them to an indicator".
The feature cost modelling objective is to provide a model which integrates the
necessary information to the definition of the product and of its manufacturing process so as
to ensure cost estimation in a preparing phase to the manufacturing.
Masmoudi, Hachicha, Bouaziz: Development Of A Welding Cost Estimation Model Based On
156
Cost parameter formulation
For a given manufacturing process, the concept feature cost allows on one hand to
estimate the direct cost corresponding to the manufacturing operation by determining for
each of them the indicator of the corresponding "feature cost".
Given "C
i
" the activity cost "I" and R
i
= R
1
, R
2
, , R
k
the whole of the resources
consumed for this activity. By definition the cost of a resource is written:
A resource cost = Inductor (number) *
k
(hour/number) * charging rate (cost/hour)
Then, a resource cost is written k:
k k i
C x C = (14)
Where
k
: resource consumption coefficient k
x
i
: activity inductor "I"
C
k
: charging rate for the resource k
The model basic equation giving an activity cost will be then the sum total costs of
different resources [3]:
k k i
R k
i
C x C
i
=
(15)
To identify the inductors, we can use several methods like the expert consulting or a
detail survey of the activity or at last, basing on a sufficient number of historical data of a
quite large number of inductors by selecting the most influent ones.
Preparing cost estimation
Each preparing feature calls for intrinsic parameters, which describe the product and
also for the position and the orientation geometric characteristics. So, the cost preparation
feature is composed of the cost machining feature, Fig. 6 (a) and of the pointing cost feature
Fig. 6 (b).
(a) The structure of the machining cost feature (b) The structure of the pointing cost feature
Figure 6: Cost feature structure.
It is quite evident that the machining cost feature has the volume of the material to
remove indicated "V
e
" as indicator. In the application of the equation (14), the machining cost
is the sum total of the resources costs which compose the machining cost feature.
ECU
OU
ECU
MA
ECU
MO
U
C C C C + + = (16)
Machining
cost feature
Volume to
remove : V
e
Machining
Cost
Machining
Labour, Tool, Machine
Pointing cost
feature
Number of
weld points :
L/L
0
Pointing
cost
Labour, Tool, Machine
Pointing
Masmoudi, Hachicha, Bouaziz: Development Of A Welding Cost Estimation Model Based On
157
where
u
C V C
u e
ECU
MO
= : is the labour cost
mu um e
ECU
MA
C V C = : is the machine utilizing cost
o
C V C
uo e
ECU
OU
= : is the lubricant and tool consumption cost
C
u
: Labour hourly cost
C
mu
: machine utilizing houry cost
C
o
: tool wear hourly cost
V
e
: volume of material to remove
um
: machine utilizing coefficient
uo
: lubricate and tool consuming coefficient.
We note "
u
" the coefficient to measure the time consuming by volume of material
removed and it is expressed in hour/cm
3
. This coefficient is given by:
v
u
D
c
= (17)
Where D
v
: the machining volume rate.
c : a complexity index ( 1 c ): we associate to each type of preparing feature
a complexity index which will be a parameter depending on the preparing technology
adopted by the enterprise. It can be determined from a statistic approach applied to each
type of feature and initialized according the experience.
To calculate the machining time for the "End to End" type preparation, we can write the
following relation:
v
U
D
L .SP c
t
= (18)
Where SP : preparing feature section of an assemblage.
L : length of the assemblage weld.
c : complexity index according to the form of preparation.
D
v
: machining volume rate.
The pointing allows a conformable disposition of the two parts to weld. It also allows to
juxtapose the borders one according to the other and after that, it controls the space all along
the weld path "L". We note "
p
" the coefficient which measures the time consumption it takes
to make a point weld which remains subject to the difficulties of execution and of
accessibility.
0
t
p
= (19)
Where : the coefficient linked to the difficulties, with 1
t
0
: the elementary time to achieve a point of weld
The resources cost of the feature "pointing cost" is principally composed of the labour
cost and the weld consuming cost for the pointing.
- The labour cost for the pointing is written:
p
0
0
C
L
L
t C
ECP
MO
= (20)
Masmoudi, Hachicha, Bouaziz: Development Of A Welding Cost Estimation Model Based On
158
Where C
p
: labour hourly cost for the pointing operation
L/L
0
: the number of points
- The pointing cost is the sum total of the resources costs, which compose the pointing
cost feature.
ECP
MA
ECP
MO
P
C C C + = (21)
We assume that the cost of one point of weld is neglect-able compared to the total
welding cost.
Welding costs estimation
Subsequent to the analytic modelling applied to the scale of a feature, according to the
relation (13) which determines the welding time, we can write:
a
a S
D
m p
t
= (22)
It is to remark that there is a harmony with the feature cost approach. In fact, if we take
the mass of materiel to be added as a convenient inductor for the welding operation, then we
can write that the time consuming coefficient of the weld noted "
s
" as follows:
a
s
D
p
(23)
After having estimated the total time of welding, it has become possible to make a
detailed cost calculating of each consumption engendered by the welding activity.
- The labour cost is:
s S a
S
MO
.C m C = (24)
Where C
s
: Labour hourly cost "UM/mn"
- The eletrode consuming cost is:
e a
S
EL
.C m C = (25)
Where C
e
: electrode cost by mass unit "UM/g"
It is clear that the consuming coefficient of the resource "Electrode" is
- The gas consuming cost is:
g a
a
S
GA
C .m
D
D
C
= (26)
Where D : protection gas density in "g of gas / cm
3
"
: the gas volumic rate in "cm
3
/mn"
C
g
: gas cost by mass unit "UM / g of gas"
- The consumption coefficient of the resource "gas" is:
a
Gaz
S
D
D
= (27)
Masmoudi, Hachicha, Bouaziz: Development Of A Welding Cost Estimation Model Based On
159
- The consumption cost of electric energy is:
k
a
a S
EE
C
D
U.I.m
C
=
60
(28)
Where I : current intensity "A"
U : the electric current tension "V"
C
k
: electric energy cost in "UM/kWh"
: the transformer average output
- The electricity consumption coefficient is:
a
EL
S
D
U.I
=
60
"kWh/g" (29)
- The utilization cost of the welding post: It corresponds to the cost engendered when the
post in function. To determine it, it is necessary to know welding machine hourly cost,
"Cms". It is expressed as follows:
ms
a
a S
Ma
C
D
m
C = (30)
The consumption coefficient of the resource (post utilization) is
a
Poste
S
D
1
= "mn/g"
The total weld cost is the sum total of the elementary costs: labour, electrode
conception, protecting gas conception, electricity consumption and welding post. It is written
as follows:
S
MA
S
EE
S
GA
S
EL
S
MO
S
C C C C C C + + + + = (31)
In any case, the raw material cost is not to neglect. It is evident that it is considered
during the welding operation.
To generalize, the total welding cost of a feature "q" is the sum total of the machining
cost, the pointing cost and of the welding cost for each feature.
P
q
U
q
S
q
T
q
C C C C + + = (32)
The total welding cost of an assemblage (i,j) is then the sum total of all the welding costs
of the feature which composes it:
=
=
k(i,j)
q
T
q
T
C (i,j) C
1
(33)
Where k (i,j): the number of assemblage (i,j) welding features.
Masmoudi, Hachicha, Bouaziz: Development Of A Welding Cost Estimation Model Based On
160
5. IMPLEMENTATION
The proposed cost estimation model for a weld assemblage based on the concept feature is
implemented to a software model for computer aided costs estimation, which is presented in
[12]. The developed software which called SOUDABASE constitutes an aided decision tool
to calculate the cost and establish the sales conditions by the experts. It provides an
informational model that integrates the information necessary to the defining of the product
and of its manufacturing process.
In order to validate the proposed model, we develop the example of Fig. 7 to show the
efficiency of SOUDABASE tool. The set for which have tried to estimate the welding cost is a
Chassis. This set is interesting in the sense that is allows us to put into evidence the cost
estimating problems, mentioned earlier.
Figure 7: The schema of the Chassis.
5.1. Proposed model estimations
The most recommended technical welding process for the Carbon steel machining is the
GMAW with a gas of (75% Ar + 25% CO2) type. The diameter of the electrode for all the
welding operation is 1.2 mm. This last parameter is according to the accessibility of the
electrode in the joint to weld. On the contrary the electrode output is determined in
concordance with welding process type. In our case the output is equal to 100%. The length
of the assemblage is determined for each type of feature which presented in Table III.
Table III: Specifications of different types of features of the chassis
Feature Type Leg Size (mm) Intensity (A) Weld length (m)
1 4,2 160 0,66
2 5,6 185 0,6
3 4,2 160 0,7
After defining the features of the Chassis, we have introduced the hourly costs of
consumption to result in costs. Thus we have associated a cost model for each assemblage.
The cost of a model presents the different unit costs (labour, gas, electrode and electricity).
The assemblage times, the consumption and the costs will be, respectively, the sum total of
the different features. These cost and time estimation are presented in Table IV.
Masmoudi, Hachicha, Bouaziz: Development Of A Welding Cost Estimation Model Based On
161
5.2. DEVISOUD estimations
In order to validate model results, we have used another estimation tool which named
DEVISOUD. It is estimation software was developed by the French centre CETIM Senlis
France (Technical centre of mechanical industries). Detailed description of DEVISOUD
software can be found in [13].
5.3. Manufacturing of Chassis
The main thing, for the manufacturing responsible of the collaborating company which
named "engineering and industrial construction" (MDI) is the total weld the total length
evaluation and the number of passes necessary for the welding costs and time estimation.
In the "Chassis" example we have studied the weld length being of 1.96 m. According to
the expertise of the company, only one pass is required to weld a 1,2 mm sheet metal. We
have calculated the total welding time of a tank which is equal to 14 min. others
measurements of different timings noted down after the welding time are indicated in Table
IV. We call back that the implementation of "SOUDABASE" results, a manufacturing and
welding time equal to 13.15 min.
Table IV: Comparative estimations and trials results
Parameters
DEVISOUD
estimation
Proposed model
estimation
Manufacturing
realisation
Weld length (m) 1.96 1.96 1.96
Amount of Welding Wire (kg) 0.22 0.19 0.20
Volume of gas used (m3) 0.08 0.07 0.07
Welding arc executing time (min) 5.05 6.6 6.0
Weld pointing time (min) 3.90 3.00 3.45
Welding time (min) 12.00 13.15 14.00
Machining time (min) N.A 0 0
Positioning time (min) 5.64 N.A 6.00
We note that the error obtained between the time calculated by our approach and the
real welding time is very small. Its average is of the size of 10 % in this case. This difference
remains small to the DEVISOUD results which show the accuracy of our calculating
approach.
6. CONCLUSION
In this paper, we have explained the concept and the basic principle peculiar to our approach
of time and cost estimating in parts welding. The model by feature which we have developed
is characterized by the weld geometric description and by the volume defining of the material
added in welding.
The formalization of the welding time is carried out by an analytic method of gas and
electrode consumption according to the feature different geometric parameters and to the
welding process. The cost estimating is assured by a parametric approach. The model
includes all the stages of the welding process: preparation and weld.
To automatize the cost generating process, it is necessary to integrate the modelling of
knowledge, the modelling of the arguments for the generating of preparation and weld
features allowing the identifications of the features cost as well as their suitable inductors.
Extending the proposed approach to this direction is our interesting research perspective.
Masmoudi, Hachicha, Bouaziz: Development Of A Welding Cost Estimation Model Based On
162
7. ACKNOWLEDGEMENTS
We would like to thank the manufacturing responsible of the collaborating company
"engineering and industrial construction" MDI for their valuable and constructive helps on the
manufacturing realisation in this study.
REFERENCES
[1] Lallemand, H. (1999). Mthodes de chiffrage et indicateurs de la fonction devis, Journal Travail et
Mthode, 545, 35-44
[2] Berlioz, M.; Martin, P.; Tichkiewitch, S. (1998). A fast and reliable cost estimation tool for Hot-
forged parts, Proceedings of the Second International Conference on Integrated Design and
Manufacturing in Mechanical Engineering, 985-992
[3] Seddiki, A.; Moisan, A.; Levaillant, G. (1995). Proposition dun Systme dAssistance
lElaboration de Devis dUsinage bas sur le concept dentit, Journal Mcanique Industrielle et
Matriax, Vol. 48, 5
[4] Duverlie, P. (1999). Mthode analogique applique aux pices usines, Journal Travail et
Mthodes, 553, 20-24
[5] Jimenez, A. (1995). Mthode paramtrique- statistique : application aux pices usines, Journal
Travail et Mthodes, 522, 19-25
[6] Hmida, F.; Martin, P.; Vernadat, F. (2005). Cost estimation in mechanical production: The Cost
Entity approach applied to integrated product engineering, International Journal Production
Economics, doi: 10.1016/ j. ijpe. 2005.02.016
[7] Bouaziz, Z.; Younes, J.B.; Zghal, A. (2002). A Fast and Reliable Tool for Estimates for Plastic
Blowing Moulds, International Journal of Advanced Manufacturing Technology, 20, 545-550
[8] Shebab, E.; Abdalla, H. (2002). An Intelligent Knowledge-Based System for Product Cost
Modelling, International Journal of Advanced Manufacturing Technology, No. 19, 49-65.
[9] Jong-Yun, J. (2002). Manufacturing Cost Estimation for Machined parts based on Manufacturing
Features, Journal of Intelligent Manufacturing, 13, 227-238
[10] Ben-Arieh, D. (2000). Cost estimation system for machined parts, International Journal of
Production Research, Vol. 1 (17), 4481-4494
[11] Ou-Yang, C.; Lin, T.S. (1997). Developing and Integrated Framework for Feature-Based Early
Manufacturing Cost Estimation, International Journal of Advanced Manufacturing Technology, 13,
618-629.
[12] Masmoudi, F.; Bouaziz, Z.; Hachicha, W. (2007). Computer-aided cost estimation of weld
operations, International Journal of Advanced Manufacturing Technology, Vol. 33, 3-4, 298-307,
doi 10.1007/s00170-006-0463-0
[13] Dossier technique de lapplication CETIM DEVISOUD ; CETIM Senlis France, Octobre 1995.