Applying Neural Networks and Analogous Estimating to Determine the Project Budget

APPLYING NEURAL
NETWORKS AND
ANALOGOUS ESTIMATING
TO DETERMINE THE
PROJECT BUDGET
Accepted for publication at
PMI Global Congress 2015 - North America
Orlando – Florida – USA – 2015

Applying Neural Networks and Analogous Estimating to Determine the Project Budget2
Abstract
This paper aims to discuss the use of the Artificial Neural Networks (ANN) to mod-
el aspects of the project budget where traditional algorithms and formulas are
not available or not easy to apply. Neural networks use a process analogous to
the human brain, where a training component takes place with existing data and
subsequently, a trained neural network becomes an “expert” in the category of
information it has been given to analyse. This “expert” can then be used to pro-
vide projections given new situations based on an adaptive learning (STERGIOU
& CIGANOS, 1996).
The article also presents a fictitious example of the use of neural networks to
determine the cost of project management activities based on the complexity,
location, budget, duration and number of relevant stakeholders. The example is
based on data from 500 projects and is used to predict the project management
cost of a given project.
Artificial Neural Networks (ANN)
Some categories of problems and challenges faced in the project environment
may depend on many subtle factors that a computer algorithm cannot be creat-
ed to calculate the results (KRIESEL, 2005). Artificial Neural Networks (ANN) are a
family of statistical learning models inspired by the way biological nervous sys-
tems, such as the brain, process information. They process records one at a time,
and“learn”by comparing their classification of the record with the known actual
classification of the record.
The errors from the initial classification of the first record are fed back into the
network, and used to modify the networks algorithm the second time around,
and so on for a large number of iterations in a learning process in order to predict
reliable results from complicated or imprecise data (STERGIOU & CIGANOS, 1996)
(Exhibit 01).
Exhibit 01 - Artificial Neural Networks Architecture (adapted from MCKIM, 1993 and
STERGIOUS & CIGANOS, 1996)
Related Podcasts
♫♫Importance of the Work
Breakdown Structure (WBS) to
the Project Cost Management
http://rvarg.as/et
♫♫Cost Estimating – Part 1 of 3
http://rvarg.as/cu
http://rvarg.as/d0
http://rvarg.as/d2
♫♫Applying Neural Networks
and Analogous Estimating to
Determine the Project Budget
http://rvarg.as/nn

ricardo-vargas.com 3
Some typical applications of ANN are
• handwriting recognition,
• stock market prediction,
• image compression,
• risk management,
• sales forecasting
• industrial process control.
The mathematical process behind the calculation uses different neural network
configurations to give the best fit to predictions. The most common network
types are briefly described below.
Probabilistic Neural Networks (PNN) – Statistical algorithm where the operations
are organized in multi-layered feedforward network with four layers (input, pat-
tern, summation and output). It is fast to be trained but it has a slow execution
and requires large memory. It is also not as general as the feedforward networks
(CHEUNG & CANNONS, 2002).
Multi-Layer Feedforward Networks (MLF) - MLF neural networks, trained with
a back-propagation learning algorithm (Exhibit 02). They are the most popular
neural networks (SVOZIL, KVASNIČKA & POSPÍCHAL, 1997).
Exhibit 02 - Training data and generalization in a Multi-Layer Feedforward Network
(SVOZIL, D , KVASNIČKA, V. & POSPÍCHAL, J. , 1997)
Generalized Regression Neural Networks (GRNN) - Closely related to PNN net-
works, it is a memory-based network that provides estimates of continuous vari-
ables. It is a one-pass learning algorithm with a highly parallel structure. The al-
gorithmic form can be used for any regression problem in which an assumption
of linearity is not justified (SPECHT, 2002).

Analogy Process and Data Set
One of the key factors of the Neural Networks is the data set used on the learning
process. If the data set is not reliable, the results from the networks calculations
will not be reliable. The use of Artificial Neural Networks can be considered one
kind of analogy (BAILER-JONES & BAILER-JONES, 2002).
Analogy is a comparison between two or more elements, typically for the pur-
pose of explanation or clarification (Exhibit 03). One of the most relevant uses
of the analogy is to forecast future results based on similar results obtained in
similar conditions (BARTHA, 2013). The challenge is to understand what a similar
condition is. Projects in the past can be a reference for future projects if the un-
derlining conditions where they were developed still exist in the project subject
to the analysis.
are to
are toas
Exhibit 03 - Simple analogy example“sock are to feet as gloves are to hands”
(Adaptedfrom Spitzig, 2013)
One of the most relevant aspects of the analogy is related to the simple process
of estimation based on similar events and facts. This process reduces the granu-
larity of all calculations, where the final project costs can be determined by a set
of fixed finite variables.
Data Set, Dependent and Independent Categories
and Numeric Variables
The first step to develop an Artificial Neural Network is to prepare the basic data
set that will be used as a reference for the “training process” of the neural net-
work. It is important to highlight that usually the right dataset is expensive and
time consuming to build (INGRASSIA & MORLINI, 2005). A dataset is composed
by a set of variables filled with information that will be used as a reference. These
references are called cases (Exhibit 04).

VARIABLES
INDEPENDENT VARIABLES DEPENDENT VARIABLE (OUTPUT)
V1 V2 V3 Vn V’1
CASES
Case 1
Case 2
Case 3
Case n
Exhibit 04 - Structure of a basic dataset
The most common variables types are
Dependent Category – dependent or output variable whose possible values are
taken from a set of possible categories; for example Yes or No, or Red, Green or
Blue.
Dependent Numeric – dependent or output variable whose possible values are
numeric.
Independent Category – an independent variable whose possible values are tak-
en from a set of possible categories; for example Yes or No, or Red, Green or Blue.
Independent Numeric – an independent variable whose possible values are nu-
meric.
In the project environment, several variables can be used to calculate the project
budget. Some common examples are
Complexity – Level of complexity of the project (Low, Medium, High). Usually it
is an independent category.
Location – Location where the project works will happen. Associated to the com-
plexity of the works and logistics. Most of the time it is an independent category.
Budget – Planned budget of the project. It is a numeric variable that can be inde-
pendent or dependent (output).
Actual Cost – Actual Expenditure of the project. It is most of the time an indepen-
dent numeric variable.
Cost Variance – The difference between the budget and the actual cost. It is a
numeric variable that can be independent or dependent (output)
Baseline Duration – Duration of the project. Independent numeric variable.
Actual Duration – Actual duration of the project. Usually an independent numer-
ic variable.

Duration Variance – The difference between the baseline duration and the actu-
al duration.
Type of Contract – Independent category variable that defines the type of the
contract used for the works in the project (ie: Fixed Firm Price, Cost Plus, Unit
Price, etc).
Number of Relevant Stakeholder Groups – Independent numeric variable that
reflect the number of relevant stakeholder groups in the project.
Some examples of input variables are presented at the Exhibit 05, 06 and 07.
INPUT VARIABLES DESCRIPTION UNIT RANGE
PWA Predominant Work Activity Category
New Construction Asphalt or
Concrete
WD Work Duration month 14–30
PW Pavement Width m 7–14
SW Shoulder Width m 0–2
GRF Ground Rise Fall nillan 2–7
ACG Average Site Clear/Grub m2
/kin 12605–30297
EWV Earthwork Volume m3
/kin 13134–31941
SURFCLASS Surface Class Category Asphalt or Concrete
BASEMATE Base Material Category Crushed Stone or Cement Stab.
OUTPUT VARIABLE
USDPERKM Unit Cost of New Construction Project
US Dollars
(2000)
572.501.64-4.006.103.95
Exhibit 05 – Example of Variables in Road Construction (SODIKOV, 2005)
DESCRIPTION RANGE
Ground floor 100–3668 m2
Area of the typical floor 0–2597 m2
No. of storeys 1–8
No. of columns 10–156
Type of foundation
1 – isolated
2 – isolated and combined
3 – raft or piles
No. of elevators 0–3
No. of rooms 2–38
Cost of structural skeleton 6,282469,680 USD
Exhibit 06 – Example of key variables for buildings (ARAFA & ALQEDRA, 2011)

PROJECT CHARACTERISTICS UNIT TYPE OF INFORMATION DESCRIPTORS
Gross Floor Area (GFA) m2
Quantitative n.a
Principal structural material No unit Categorical
1 – steel
2 – concrete
Procurement route No unit Categorical
1 – traditional
2 – design and construct
Type of work No unit Categorical
1 – residential
2 – commercial
3 – office
Location No unit Categorical
1 – central business district
2 – metropolitan
3 – regional
Sector No unit Categorical
1 – private sector
2 – public sector
Estimating method No unit Categorical
1 – superficial method
2 – approximate quantities
Number of storey No unit Categorical
1 – one to two storey(s)
2 – three to seven storeys
3 – eight storeys and above
Estimated Sum Cost/m2
Quantitative n.a
Exhibit 07 – Example of variables for a building construction ( AIBINU, DASSANAYAKE & THIEN, 2011)
Training Artificial Neural Networks
When the dataset is ready the network is ready to be trained. Two approaches
can be used for the learning process: supervised or adaptive training.
In the supervised training, both inputs and outputs are provided and the net-
work compares the results with the provided output. This allows the monitoring
of how well an artificial neural network is converging on the ability to predict the
right answer.
For the adaptive training, only the inputs are provided. Using self-organization
mechanisms, the neural networks benefits from continuous learning in order
to face new situations and environments. This kind of network is usually called
self-organizing map (SOM) and was developed by Teuvo Kohonen (KOHONEN,
2014).
One of the biggest challenges of the training method is to decide on which net-
work to use and the runtime process in the computer. Some networks can be
trained in seconds but in some complex cases with several variables and cases,
hours can be needed just for the training process.

The results of the training process are complex formulas that relate the input or
independent variables with the outputs (dependable variables) like the graph
presented in the Exhibit 2.
Most of the commercial software packages usually test the results of the training
with some data points to evaluate the quality of the training. Around 10 to 20%
of the sample is used for testing purposes (Exhibit 08).
Exhibit 08 – Training results example to forecast the bloody pressure where some data is
used for testing the network results (Palisade Neural Tools software example)
Prediction Results
After the training, the model is ready to predict future results. The most relevant
information that should be a focus of investigation is the contribution of each
individual variable to the predicted results (Exhibit 09) and the reliability of the
model (Exhibit 10).

Exhibit 09 - Example of Relative Variable impacts, demonstrating that the Salary variable is
responsible for more than 50% of the impact in the dependent variable (Palisade Neural
Tools software example)
Exhibit 10 – Example of histogram of Probability of Incorrect Categories showing a
chance of 30% that 5% of the prediction can be wrong (Palisade Neural Tools software
example)
It is important to highlight that one trained network that fails to get a reliable
result in 30% of the cases is much more unreliable than another one that fails in
only 1% of the cases.

Example of Cost Modeling using Artificial Neural
Networks
In order to exemplify the process, a fictitious example was developed to predict
the project management costs on historical data provided by 500 cases.The vari-
ables used are described in the Exhibit 11.
NAME DESCRIPTION VARIABLE TYPE
Project ID ID Count of each project in the dataset –
Location Location where the project was developed (local or remote sites) Independent Category
Complexity Qualitative level of project complexity (Low, Medium and High) Independent Category
Budget Project Budget (between $500,000 and $2,000,000) Independent Numeric
Duration Project Duration (Between 12 and 36 months) Independent Numeric
Relevant Stakeholder
Groups
Number of relevant stakeholder groups for communication and
monitoring (between 3 and 5)
Independent Category
PM Cost
Actual cost of the project management activities (planning,
budgeting, controlling)
Dependent Numeric
(Output)
Exhibit 11 – Variables used on the example dataset
The profiles of the cases used for the training are presented at the Exhibit 12, 13,
14, 15 and 16 and the full dataset is presented in the Appendix.
Location Projects %
Local 250 50.00%
Remote 250 50.00%
Total 500 100.00%
Local
50%
Remote
50%
Exhibit 12 – Distribution of cases by Location
Complexity Projects %
Low 201 40.20%
Medium 176 35.20%
High 123 24.60%
Total 500 100.00%
Low

40%

Medium
35%

High

25%

Exhibit 13 – Distribution of cases by Complexity

Budget Range Projects % Budget Range Projects
500000-750000 158 31.60% 500000-750000 158
750000-1000000 111 22.20% 750000-1000000 111
1000000-1250000 99 19.80% 1000000-1250000 99
1250000-1500000 74 14.80% 1250000-1500000 74
1500000-1750000 43 8.60% 1500000-1750000 43
1750000-2000000 15 3.00% 1750000-2000000 15
Total 500 100.00% Total 5000
20
40
60
80
100
120
140
160
180
Exhibit 14 – Distribution of cases by Project Budget
Duration Range Projects % Duration Range Projects
12-17 145 29.00% 12-17 145
18-23 189 37.80% 18-23 189
24-29 96 19.20% 24-29 96
30-36 70 14.00% 30-36 70
Total 500 100.00% Total 500
0
20
40
60
80
100
120
140
160
180
200
12-17 18-23 24-29 30-36
Exhibit 15 – Distribution of cases by Project Duration
Stakeholder Groups Projects %
3 238 47.60%
4 158 31.60%
5 104 20.80%
Total 500 100.00%
3
47%
4
32%
5
21%
Exhibit 16 – Distribution of cases by Relevant Stakeholder Groups
The training and the tests were executed using the software Palisade Neural
Tools.The test was executed in 20% of the sample and a GRNN Numeric Predictor.
The summary of the training of the ANN is presented at the Exhibit 17.

NET INFORMATION
Name Net Trained on Project Data
Configuration GRNN Numeric Predictor
Location This Workbook
Independent Category Variables
3 (Location, Complexity,
Relevant Stakeholder Groups)
Independent Numeric Variables 2 (Budget, Duration)
Dependent Variable Numeric Var. (PM Cost)
TESTING
Number of Cases 100
% Bad Predictions (30% Tolerance) 0.0000%
Root Mean Square Error 1,508.66
Mean Absolute Error 1,170.24
Std. Deviation of Abs. Error 952.16
TRAINING
Number of Cases 400
Training Time 00:00:38
Number of Trials 75
Reason Stopped Auto-Stopped
% Bad Predictions (30% Tolerance) 0.0000%
Root Mean Square Error 944.70
Mean Absolute Error 706.47
Std. Deviation of Abs. Error 627.18
PREDICTION
Number of Cases 1
Live Prediction Enabled YES
DATA SET
Name Project Data
Number of Rows 501
Manual Case Tags NO
VARIABLE IMPACT ANALYSIS
Budget 54.3124%
Duration 25.4821%
Location 8.2800%
Complexity 7.0677%
Relevant Stakeholder Groups 4.8578%
Exhibit 17 – Palisade Neural Tools Summary Table
The relative impact of the five independent variables are described at the Exhibit
18, demonstrating that more than 50% of the impact in the Project Management
cost is related to the project budget in this fictitious example.
10000

20000

30000

40000

50000

60000

Predicted

54.3124%

25.4821%

8.2800%

7.0677%

4.8578%

0%
10%
20%
30%
40%
50%
60%
70%
80%

Budget

Dura9on

Loca9on

Complexity

Relevant
Stakeholder
Groups

Rela%ve
Variable
Impacts

Exhibit 18 – Relative Variable Impacts

The training and tests were used to predict the Project Management Cost of a
fictitious project with the following variables
NAME VARIABLE TYPE
Location Local Project
Complexity High Complexity
Budget $810,756
Duration 18 months
Relevant Stakeholder Groups 5 Stakeholder groups
Relevant Stakeholder Groups Independent Category
PM Cost Dependent Numeric (Output)
Exhibit 19 – Basic information of a future project to be used to predict the
Project Management costs
After running the simulation, the Project Management cost predictions based on
the patterns in the known data is $24,344.75, approximately 3% of the project
budget.
Conclusions
The use of Artificial Neural Networks can be a helpful tool to determine aspects of
the project budget like the cost of project management, the estimated bid value
of a supplier or the insurance cost of equipment. The Neural Networks allows
some precise decision making process without an algorithm or a formula based
process.
With the recent development of software tools, the calculation process becomes
very simple and straightforward. However, the biggest challenge to produce re-
liable results lies in the quality of the known information. The whole process is
based on actual results, and most of the time the most expensive and laborious
part of the process is related to getting enough reliable data to train and test the
process.
References
AIBINU, A. A., DASSANAYAKE, D. &THIEN,V. C. (2011). Use of Artificial Intelligence to
Predict the Accuracy of Pretender Building Cost Estimate. Amsterdam: Management
and Innovation for a Sustainable Built Environment.
ARAFA, M. & ALQEDRA, M. (2011). Early Stage Cost Estimation of Buildings Con-
struction Projects using Artificial Neural Networks. Faisalabad: Journal of Artificial
Intelligence.

BAILER-JONES, D & BAILER-JONES, C. (2002). Modeling data: Analogies in neural
networks, simulated annealing and genetic algorithms. New York: Model-Based
Reasoning: Science, Technology, Values/Kluwer Academic/Plenum Publishers.
BARTHA, P (2013). Analogy and Analogical Reasoning. Palo Alto: Stanford Center
for the Study of Language and Information.
CHEUNG, V. & CANNONS, K. (2002). An Introduction to Neural Networks. Winnipeg,
University of Manitoba.
INGRASSIA, S & MORLINI, I (2005). Neural Network Modeling for Small Datasets In
Technometrics: Vol 47, n 3. Alexandria: American Statistical Association and the
American Society for Quality
KOHONEN, T. (2014). MATLAB Implementations and Applications of the Self-Orga-
nizing Map. Helsinki: Aalto University, School of Science.
KRIESEL, D. (2005). A Brief Introduction to Neural Networks. Downloaded on
07/01/2015 at http://www.dkriesel.com/_media/science/neuronalenetze-en-ze-
ta2-2col-dkrieselcom.pdf
MCKIM, R. A. (1993). Neural Network Applications for Project Management. New-
town Square: Project Management Journal.
SODIKOV, J. (2005). Cost Estimation of Highway Projects in Developing Countries:
Artificial Neural Network Approach. Tokyo: Journal of the Eastern Asia Society for
Transportation Studies, Vol. 6.
SPECHT, D. F. (2002). A General Regression Neural Network. New York, IEEE Transac-
tions on Neural Networks, Vol 2, Issue 6.
SPITZIG, S. (2013). Analogy in Literature: Definition & Examples in SAT Prep: Help and
Review. Link accessed on 06/30/2015: http://study.com/academy/lesson/analo-
gy-in-literature-definition-examples-quiz.html
STERGIOUS, C & CIGANOS, D. (1996). NeuralNetworksinSurpriseJournalVol4,n11.
London, Imperial College London.
SVOZIL, D, KVASNIČKA, V. & POSPÍCHAL, J. (1997). Introduction to multi-layer
feed-forward neural networks In Chemometrics and Intelligent Laboratory Systems,
Vol 39. Amsterdam, Elsevier Journals.

Appendix – Example Dataset
LOCA-
TION
COMPLE-
XITY BUDGET
DURA-
TION
RELE-
VANT
STAKE-
HOLDER
GROUPS PM COST
% PM
COST
TAG
USED TEST
GOOD
OR BAD RESIDUAL %
Remote Medium 703,998.33 17 3 21,547.87 3.06% train
Remote High 902,327.29 17 5 33,934.58 3.76% test 33,275.83 Good 658.75 1.94%
Local Low 904,824.77 27 3 14,789.98 1.63% train
Local Low 640,833.02 17 3 15,128.69 2.36% train
Local Low 683,992.89 16 3 16,985.82 2.48% test 17,240.60 Good -254.78 -1.50%
Remote High 1,467,802.93 20 5 49,416.03 3.37% train
Remote High 569,532.07 26 4 16,983.74 2.98% train
Remote Low 1,235,140.98 12 4 47,896.02 3.88% train
Remote Low 1,254,182.69 13 4 45,954.54 3.66% test 43,510.55 Good 2,443.99 5.32%
Local High 1,045,689.31 15 3 31,603.05 3.02% test 29,729.94 Good 1,873.11 5.93%
Local Medium 1,070,909.21 20 5 27,486.67 2.57% train
Remote Low 600,174.43 14 4 21,491.96 3.58% train
Remote Low 1,333,972.58 13 5 50,212.10 3.76% train
Local Low 1,618,395.90 12 3 49,810.63 3.08% test 43,252.94 Good 6,557.69 13.17%
Remote Low 580,524.22 15 3 18,125.26 3.12% test 18,178.96 Good -53.70 -0.30%
Remote Low 1,277,669.74 26 4 30,434.75 2.38% train
Local High 1,465,538.27 27 5 35,679.52 2.43% test 36,732.46 Good -1,052.93 -2.95%
Local High 534,389.92 19 5 16,322.33 3.05% test 16,106.84 Good 215.49 1.32%
Remote Low 938,755.52 14 4 33,616.39 3.58% train
Local Medium 765,884.98 16 3 20,551.25 2.68% train
Local Low 762,219.86 16 3 18,928.46 2.48% train
Local Low 964,410.00 20 3 19,931.14 2.07% train
Remote Low 911,404.26 24 4 23,595.24 2.59% test 22,578.66 Good 1,016.59 4.31%
Local Low 1,126,200.40 21 3 21,254.80 1.89% test 22,151.95 Good -897.15 -4.22%
Local High 708,383.15 21 3 17,619.63 2.49% train
Local Low 852,403.45 17 4 22,680.62 2.66% train
Remote Low 816,178.39 16 3 24,349.32 2.98% train

LOCA-
TION
COMPLE-
XITY BUDGET
DURA-
TION
RELE-
VANT
STAKE-
HOLDER
GROUPS PM COST
% PM
COST
TAG
USED TEST
GOOD
OR BAD RESIDUAL %
Remote Low 1,151,686.39 22 5 31,270.03 2.72% train
Local Low 1,219,803.85 21 5 27,900.59 2.29% train
Remote Low 1,359,202.77 17 4 41,602.27 3.06% train
Local Low 693,228.75 15 3 18,178.00 2.62% train
Local High 801,510.16 28 3 16,755.38 2.09% train
Local High 511,096.39 26 5 13,196.77 2.58% train
Local Low 590,242.71 15 5 17,838.45 3.02% train
Remote Low 1,116,386.68 17 4 34,170.19 3.06% train
Remote Medium 1,123,846.83 23 4 30,897.64 2.75% test 31,578.29 Good -680.64 -2.20%
Local Low 894,732.93 19 3 19,276.00 2.15% train
Remote Low 1,171,008.92 19 4 33,425.11 2.85% train
Remote Low 551,582.86 21 4 15,374.28 2.79% train
Local Low 1,323,310.37 28 3 19,723.63 1.49% train
Remote Low 845,707.18 20 3 21,706.48 2.57% train
Local Low 1,399,151.60 12 4 47,260.23 3.38% train
Remote Low 1,629,835.05 19 4 46,521.78 2.85% train
Local Low 1,747,728.47 19 5 42,896.00 2.45% train
Local High 584,824.62 24 4 15,140.46 2.59% train
Remote Low 1,234,685.15 19 4 35,242.68 2.85% train
Local Low 982,920.06 21 3 19,533.59 1.99% train
Local High 1,082,133.01 21 4 29,080.18 2.69% test 28,236.53 Good 843.65 2.90%
Remote High 1,035,386.38 25 4 30,371.33 2.93% test 31,220.71 Good -849.38 -2.80%
Remote Medium 1,264,034.73 14 3 42,736.41 3.38% train
Remote Low 1,002,553.31 13 5 37,737.13 3.76% train
Local Low 1,420,828.51 19 3 29,189.30 2.05% train
Local Low 1,709,337.52 15 4 48,241.30 2.82% train
Local High 609,335.11 28 4 14,566.01 2.39% test 14,651.91 Good -85.90 -0.59%

LOCA-
TION
COMPLE-
XITY BUDGET
DURA-
TION
RELE-
VANT
STAKE-
HOLDER
GROUPS PM COST
% PM
COST
TAG
USED TEST
GOOD
OR BAD RESIDUAL %
Remote Low 1,297,801.29 23 3 29,191.12 2.25% train
Remote Low 1,119,369.76 14 3 35,606.62 3.18% train
Local Low 925,628.02 19 4 22,718.48 2.45% train
Local High 667,414.59 24 3 15,276.38 2.29% train
Local Low 951,195.05 23 5 21,395.00 2.25% test 20,650.89 Good 744.11 3.48%
Local Low 1,363,830.91 18 5 34,802.94 2.55% train
Local High 1,125,818.31 30 5 26,018.91 2.31% train
Local Medium 555,745.83 16 3 14,912.51 2.68% test 17,164.08 Good -2,251.57 -15.10%
Local High 1,437,619.16 15 5 49,198.52 3.42% train
Remote Low 512,839.97 15 3 16,012.00 3.12% train
Remote Low 1,108,388.88 18 3 29,392.83 2.65% train
Local Low 1,491,757.71 14 4 44,468.59 2.98% train
Local High 573,367.88 25 4 14,525.32 2.53% train
Remote Low 1,340,923.44 30 3 25,626.54 1.91% train
Remote Low 918,511.12 15 3 28,677.96 3.12% train
Local Low 823,502.63 21 3 16,365.48 1.99% train
Local Low 1,277,239.09 22 5 28,292.78 2.22% train
Remote Low 615,510.45 19 5 18,800.06 3.05% train
Local Low 852,551.98 24 3 15,251.21 1.79% test 16,513.71 Good -1,262.50 -8.28%
Local Low 514,229.05 22 5 11,905.18 2.32% train
Remote High 813,527.00 16 4 30,778.44 3.78% test 31,776.14 Good -997.71 -3.24%
Local Low 534,936.99 27 5 10,883.66 2.03% test 11,947.44 Good -1,063.79 -9.77%
Local High 968,941.49 20 4 27,776.32 2.87% train
Local High 1,455,430.69 23 3 32,736.64 2.25% train
Remote Low 553,402.62 20 4 15,864.21 2.87% train
Remote Low 1,550,217.54 15 3 46,851.02 3.02% train
Local Low 958,266.50 21 4 21,918.44 2.29% train

LOCA-
TION
COMPLE-
XITY BUDGET
DURA-
TION
RELE-
VANT
STAKE-
HOLDER
GROUPS PM COST
% PM
COST
TAG
USED TEST
GOOD
OR BAD RESIDUAL %
Remote Low 1,572,775.22 24 3 34,426.30 2.19% train
Local Low 1,286,047.40 13 5 41,977.91 3.26% test 43,382.91 Good -1,405.01 -3.35%
Local Medium 506,773.64 28 3 9,073.66 1.79% test 10,009.00 Good -935.33 -10.31%
Remote Medium 1,561,191.51 16 5 54,381.50 3.48% test 53,120.34 Good 1,261.17 2.32%
Local High 903,316.02 20 4 25,895.06 2.87% train
Remote Medium 580,211.77 20 3 16,052.53 2.77% test 17,675.50 Good -1,622.97 -10.11%
Remote Low 595,520.47 15 5 20,975.55 3.52% train
Local Low 655,421.89 16 5 18,898.00 2.88% train
Local High 897,256.60 20 4 25,721.36 2.87% train
Remote Low 604,357.31 19 3 16,041.98 2.65% train
Local Low 504,023.79 19 3 10,858.62 2.15% train
Remote Low 914,671.35 20 3 23,476.56 2.57% test 23,342.33 Good 134.24 0.57%
Local Low 816,984.05 33 3 11,520.30 1.41% train
Local Medium 1,102,518.04 15 5 34,423.06 3.12% test 32,270.39 Good 2,152.67 6.25%
Local Low 866,386.50 27 4 16,760.84 1.93% train
Remote Low 945,814.91 19 3 25,105.58 2.65% train
Remote Low 1,007,543.31 21 3 24,053.10 2.39% train
Local Medium 1,063,937.52 33 3 16,066.53 1.51% test 19,098.46 Good -3,031.93 -18.87%
Remote Low 1,316,509.72 17 3 36,345.99 2.76% train
Local Low 819,992.37 36 5 14,152.46 1.73% train
Local Low 556,860.84 22 3 10,664.73 1.92% train
Remote Low 1,128,949.92 36 3 19,484.84 1.73% train
Remote Low 1,140,022.19 16 3 32,870.64 2.88% train
Local Medium 1,277,998.06 23 5 30,023.69 2.35% test 29,794.51 Good 229.18 0.76%
Local Low 1,370,381.07 13 4 43,360.26 3.16% train

LOCA-
TION
COMPLE-
XITY BUDGET
DURA-
TION
RELE-
VANT
STAKE-
HOLDER
GROUPS PM COST
% PM
COST
TAG
USED TEST
GOOD
OR BAD RESIDUAL %
Local Low 606,852.57 26 5 12,634.98 2.08% train
Remote Low 951,616.00 12 5 38,804.79 4.08% train
Remote Low 580,202.08 32 3 11,265.59 1.94% train
Local Low 1,283,482.92 30 5 23,245.30 1.81% train
Remote Low 1,226,103.02 27 3 24,945.90 2.03% train
Remote Low 514,184.61 22 3 12,418.34 2.42% train
Local Low 894,012.12 17 5 24,681.75 2.76% train
Local Low 535,711.70 22 4 11,866.83 2.22% train
Local High 1,770,395.16 22 3 40,987.33 2.32% train
Local High 1,405,512.56 16 3 40,525.61 2.88% train
Local Low 1,103,463.05 15 3 27,831.79 2.52% train
Remote Medium 885,202.32 20 4 27,146.20 3.07% test 25,689.17 Good 1,457.04 5.37%
Local Low 1,220,977.54 27 4 22,399.66 1.83% train
Remote Low 812,827.47 19 3 21,575.58 2.65% train
Remote Low 800,720.74 20 4 22,953.99 2.87% train
Local Low 1,360,528.32 31 4 22,792.51 1.68% test 22,768.65 Good 23.86 0.10%
Local Medium 1,048,802.19 22 4 24,281.36 2.32% test 25,044.15 Good -762.79 -3.14%
Local High 964,150.49 20 4 27,638.98 2.87% train
Remote Low 1,270,776.17 21 4 34,149.59 2.69% train
Local Medium 1,236,912.47 26 5 26,990.06 2.18% test 27,084.03 Good -93.96 -0.35%

LOCA-
TION
COMPLE-
XITY BUDGET
DURA-
TION
RELE-
VANT
STAKE-
HOLDER
GROUPS PM COST
% PM
COST
TAG
USED TEST
GOOD
OR BAD RESIDUAL %
Remote Low 540,059.74 12 3 19,862.20 3.68% train
Local Low 704,793.42 16 3 17,502.37 2.48% train
Local Low 1,325,031.76 28 4 23,724.38 1.79% train
Remote Low 1,167,617.40 16 4 37,169.15 3.18% train
Local High 1,192,348.18 26 3 24,825.30 2.08% train
Remote Low 531,703.85 15 3 16,600.98 3.12% test 17,783.95 Good -1,182.98 -7.13%
Local High 1,830,390.71 28 4 37,189.23 2.03% train
Local Low 948,143.98 15 3 24,862.44 2.62% train
Remote High 1,657,808.50 31 4 44,350.83 2.68% test 46,338.40 Good -1,987.57 -4.48%
Local High 1,767,884.50 25 3 37,714.87 2.13% train
Local High 771,280.77 21 3 19,184.08 2.49% train
Remote Low 625,360.95 26 5 16,147.14 2.58% train
Remote Low 599,119.11 15 5 21,102.31 3.52% train

LOCA-
TION
COMPLE-
XITY BUDGET
DURA-
TION
RELE-
VANT
STAKE-
HOLDER
GROUPS PM COST
% PM
COST
TAG
USED TEST
GOOD
OR BAD RESIDUAL %
Local High 854,967.79 22 3 20,648.77 2.42% train
Local Low 614,910.44 18 4 15,691.60 2.55% train
Local High 1,010,812.89 16 3 29,145.11 2.88% train
Local Low 1,605,359.49 15 3 40,490.73 2.52% train
Local Low 909,185.58 25 5 19,395.96 2.13% train
Remote Low 1,750,698.16 16 5 57,481.26 3.28% train
Remote Low 777,901.92 18 3 21,406.71 2.75% train
Local High 785,613.65 34 3 14,772.62 1.88% train
Local Low 706,311.75 19 3 15,216.68 2.15% train
Remote Low 739,540.41 21 5 21,352.76 2.89% train
Local High 1,201,962.84 21 3 28,694.48 2.39% train
Local High 1,050,306.81 16 4 33,434.77 3.18% train
Local Low 888,601.69 20 3 18,364.43 2.07% train
Local High 602,631.37 25 3 13,458.77 2.23% train
Remote Low 615,606.68 31 5 14,622.31 2.38% train
Local Low 1,038,350.37 30 3 14,652.28 1.41% train
Remote Low 1,008,605.83 15 3 30,482.31 3.02% train
Local High 1,033,174.25 18 4 30,497.77 2.95% test 28,677.64 Good 1,820.13 5.97%
Remote Low 658,752.67 20 3 16,907.99 2.57% train

LOCA-
TION
COMPLE-
XITY BUDGET
DURA-
TION
RELE-
VANT
STAKE-
HOLDER
GROUPS PM COST
% PM
COST
TAG
USED TEST
GOOD
OR BAD RESIDUAL %
Remote Low 556,260.00 14 4 19,919.41 3.58% test 20,022.91 Good -103.50 -0.52%
Remote Low 1,262,421.26 18 5 38,527.23 3.05% test 37,941.67 Good 585.56 1.52%
Local Low 537,807.95 25 4 10,935.43 2.03% train
Local Low 955,684.84 26 3 16,075.11 1.68% train
Remote Low 995,130.04 27 3 21,241.73 2.13% train
Remote Low 1,141,834.26 19 3 29,166.85 2.55% train
Local Low 1,149,263.50 28 5 21,726.55 1.89% test 22,898.96 Good -1,172.40 -5.40%
Local High 697,640.20 21 3 17,352.42 2.49% train
Local High 1,439,365.41 15 3 43,500.82 3.02% train
Local High 644,115.39 35 5 14,507.93 2.25% train
Remote Low 875,107.31 19 3 23,228.73 2.65% train
Local Low 1,471,608.90 30 5 26,652.47 1.81% train
Local Low 1,515,009.77 24 3 25,586.83 1.69% train
Remote Low 540,092.44 20 3 13,862.37 2.57% train
Remote Medium 1,506,018.23 27 3 33,653.00 2.23% test 36,646.12 Good -2,993.12 -8.89%
Local Low 1,223,357.49 20 3 24,059.36 1.97% train
Remote Low 1,177,260.33 22 3 27,255.36 2.32% train
Remote Low 567,631.63 20 3 14,569.21 2.57% train
Remote Low 1,059,977.22 30 3 20,257.34 1.91% train
Remote Low 1,242,074.92 22 3 28,755.92 2.32% train
Remote Low 533,466.84 19 5 16,294.14 3.05% train

LOCA-
TION
COMPLE-
XITY BUDGET
DURA-
TION
RELE-
VANT
STAKE-
HOLDER
GROUPS PM COST
% PM
COST
TAG
USED TEST
GOOD
OR BAD RESIDUAL %
Local High 557,365.65 17 4 17,617.13 3.16% train
Local Low 575,092.91 30 3 8,690.29 1.51% train
Local Low 1,283,649.31 21 4 28,077.28 2.19% train
Local High 1,127,308.76 21 4 30,294.19 2.69% train
Remote Low 863,172.40 15 5 30,402.85 3.52% train
Remote Low 562,812.48 15 4 19,260.69 3.42% train
Remote Low 518,239.43 34 5 11,817.89 2.28% train
Local High 1,282,007.44 15 4 42,591.14 3.32% train
Local Medium 838,064.15 24 3 16,668.16 1.99% test 18,249.56 Good -1,581.39 -9.49%
Local Low 673,022.37 15 3 17,648.14 2.62% train
Local High 1,768,637.41 33 5 39,088.67 2.21% train
Local Low 1,721,279.85 20 3 33,851.84 1.97% train
Remote Low 1,259,321.65 18 5 38,432.63 3.05% train
Local High 753,129.64 34 3 14,161.79 1.88% test 14,469.25 Good -307.46 -2.17%
Local High 1,591,469.31 13 4 58,313.07 3.66% train
Local High 1,815,026.04 15 5 62,114.22 3.42% train
Remote Medium 539,535.55 16 4 18,793.82 3.48% test 19,527.43 Good -733.61 -3.90%
Local Low 1,191,074.21 24 3 20,115.92 1.69% train
Local Low 866,382.17 26 3 14,572.99 1.68% train
Local Low 1,200,895.52 19 4 28,273.72 2.35% train

LOCA-
TION
COMPLE-
XITY BUDGET
DURA-
TION
RELE-
VANT
STAKE-
HOLDER
GROUPS PM COST
% PM
COST
TAG
USED TEST
GOOD
OR BAD RESIDUAL %
Local High 708,604.12 15 3 22,124.20 3.12% train
Remote High 721,495.49 21 5 24,439.23 3.39% test 22,337.15 Good 2,102.08 8.60%
Remote Low 1,587,492.55 16 5 52,122.67 3.28% train
Local Low 879,426.14 20 3 18,174.81 2.07% train
Remote Low 814,569.24 24 3 18,644.58 2.29% train
Remote Low 550,677.71 29 3 11,285.73 2.05% train
Remote Low 654,244.47 28 3 13,676.82 2.09% train
Remote Low 633,649.59 26 3 13,826.56 2.18% train
Remote Low 641,383.96 35 3 11,880.87 1.85% train
Remote Medium 1,481,728.02 17 3 43,870.77 2.96% test 42,949.19 Good 921.58 2.10%
Remote Low 1,827,161.32 16 3 52,683.15 2.88% train
Remote Low 964,830.02 15 4 33,018.63 3.42% train
Local Low 632,906.70 22 3 12,121.12 1.92% train
Local Low 526,027.21 34 3 7,261.24 1.38% train
Local Low 886,535.29 20 3 18,321.73 2.07% train
Local Low 1,248,941.20 33 3 16,362.39 1.31% train
Local High 618,518.54 21 3 15,384.42 2.49% train
Local Low 530,522.61 15 3 13,911.48 2.62% train
Local Low 1,009,146.22 22 4 21,344.97 2.12% train
Local High 1,021,674.63 27 3 20,786.66 2.03% train
Remote Low 1,121,376.46 22 5 30,447.07 2.72% train
Local Low 638,464.57 31 4 11,334.46 1.78% train
Local Medium 987,396.42 15 3 27,866.52 2.82% test 26,778.88 Good 1,087.64 3.90%
Local High 694,023.76 29 4 16,305.57 2.35% train

LOCA-
TION
COMPLE-
XITY BUDGET
DURA-
TION
RELE-
VANT
STAKE-
HOLDER
GROUPS PM COST
% PM
COST
TAG
USED TEST
GOOD
OR BAD RESIDUAL %
Local Low 1,633,982.66 35 4 25,365.64 1.55% train
Remote Low 573,443.71 23 3 13,471.77 2.35% train
Local Medium 586,644.29 30 5 12,384.71 2.11% test 12,332.61 Good 52.10 0.42%
Local Low 1,194,191.56 20 3 23,485.77 1.97% train
Local Low 640,851.46 29 4 11,852.07 1.85% train
Local High 653,026.72 15 3 20,388.95 3.12% train
Remote Low 1,123,457.64 30 3 21,470.52 1.91% train
Remote Low 1,367,692.12 17 4 41,862.11 3.06% train
Local Low 573,685.93 27 4 11,098.34 1.93% train
Local Low 1,022,032.17 16 3 24,358.43 2.38% test 24,273.38 Good 85.06 0.35%
Local High 1,680,626.26 17 5 53,120.97 3.16% train
Remote High 1,556,942.09 17 3 50,768.52 3.26% test 52,653.72 Good -1,885.20 -3.71%
Local Low 1,368,257.92 16 4 36,714.92 2.68% train
Remote Low 551,086.24 20 3 14,144.55 2.57% train
Local High 584,003.61 16 4 19,174.79 3.28% train
Remote Low 625,248.12 18 3 17,205.90 2.75% train
Remote Low 627,751.93 15 4 21,483.07 3.42% test 20,197.79 Good 1,285.28 5.98%
Local High 1,268,675.38 20 4 35,100.02 2.77% train
Local High 1,547,340.00 27 5 37,671.04 2.43% train
Remote High 1,185,245.00 32 4 31,310.22 2.64% test 28,602.72 Good 2,707.50 8.65%
Local Low 524,430.06 33 3 7,394.99 1.41% train

LOCA-
TION
COMPLE-
XITY BUDGET
DURA-
TION
RELE-
VANT
STAKE-
HOLDER
GROUPS PM COST
% PM
COST
TAG
USED TEST
GOOD
OR BAD RESIDUAL %
Remote Medium 856,258.79 24 3 21,311.33 2.49% test 21,521.34 Good -210.01 -0.99%
Local High 620,702.06 31 4 14,122.64 2.28% train
Remote Medium 1,284,054.66 24 4 34,526.80 2.69% test 33,189.13 Good 1,337.68 3.87%
Remote Low 1,221,724.86 15 5 41,810.14 3.42% train
Local Low 1,713,664.43 16 4 45,983.33 2.68% train
Remote Medium 1,277,241.72 36 3 24,598.73 1.93% test 25,616.68 Good -1,017.95 -4.14%
Local Low 889,253.80 16 4 24,750.90 2.78% test 22,245.27 Good 2,505.62 10.12%
Local Low 510,489.36 31 4 9,062.56 1.78% train
Remote Medium 843,268.82 20 3 23,330.44 2.77% test 23,204.29 Good 126.15 0.54%
Local High 1,296,821.96 23 5 34,356.38 2.65% train
Local Low 1,066,325.50 20 3 20,971.07 1.97% train
Remote Low 941,441.39 16 5 31,852.10 3.38% train
Local Low 546,951.49 23 3 10,114.64 1.85% train
Remote Low 661,069.81 20 4 18,950.67 2.87% train
Remote Low 1,240,392.01 16 3 35,764.64 2.88% train
Remote Low 869,608.71 23 4 23,038.33 2.65% train
Remote Low 568,098.78 22 3 13,720.45 2.42% train
Remote High 809,584.87 21 5 27,423.08 3.39% test 28,005.36 Good -582.28 -2.12%
Local Low 1,565,873.75 23 5 33,654.94 2.15% train
Remote Low 686,957.91 19 3 18,234.51 2.65% train

Applying Neural Networks and Analogous Estimating to Determine the Project Budget

Applying Neural Networks and Analogous Estimating to Determine the Project Budget

More Related Content

What's hot (20)

Viewers also liked (17)

Similar to Applying Neural Networks and Analogous Estimating to Determine the Project Budget (20)

More from Ricardo Viana Vargas (17)

Recently uploaded (20)

Applying Neural Networks and Analogous Estimating to Determine the Project Budget