Global Journal on

Technology
Vol 5 (2014) 01-07

Selected Paper of 4th World Conference on Information Technology (WCIT-2013)

REACH 4 P3 – Risk Evaluation Analysis & Classification for
Hierarchical Process & Project Plans
Zsolt Tibor Kosztyán*, University of Pannonia, Department of Quantitative Methods, Veszprém, Hungary.

Suggested Citation:
Kosztyán, Zs.T. REACH 4 P3 – Risk Evaluation Analysis & Classification for Hierarchical Process & Project Plans,
Global Journal on Technology [Online]. 2014, 05, pp 01-07. Available from: www.awer-center.org/pitcs

Received April 20, 2013; revised July 05, 2013; accepted October 03, 2013.
Selection and peer review under responsibility of Prof. Dr. Hafize Keser.
©2014 SPROC - Academic World Education & Research Center. All rights reserved.

Abstract

In former research works matrix-based methods were developed for supporting multilevel project-planning
problems. By using the introduced method traditional agile and extreme project management approaches can
also be supported. Best project scenarios can be selected by genetic algorithms, and also by exact algorithms.
However these methods are used for project planning, they can also be used for simulations, and can be applied
for risk management purposes. Score values can be attached to the tasks, dependencies and also to the
subprojects. In this way score value of a project scenario and a possible project plan can be described by a
function of the score values of task completion and/or subproject completions. In this model scope of the
project can be characterized as mandatory and score values of compulsory task completions (see i.e. MoSCoW-
analysis for IT-projects). The goal of this research is to model the project success in case of different types of
projects and project management approaches. This paper is the first one in a series in matrix-based project risk
management. The risks of two kinds of project management are compared.

Keywords: matrix-based risk planning, risk analysis and evaluation in project management, framework for risk
management.

* ADDRESS FOR CORRESPONDENCE: Zsolt Tibor Kosztyán, University of Pannonia, Department of Quantitative Methods, Egyetem
str. 10., H-8200 Veszprém, Hungary, E-mail address: [email protected] / Tel.: +36-88-624-000
Kosztyán, Zs.T. REACH 4 P3 – Risk Evaluation Analysis & Classification for Hierarchical Process & Project Plans, Global Journal on Technology
[Online]. 2014, 05, pp 01-07. Available from: www.awer-center.org/pitcs

1. Introduction
There are a lot of recent researches about project success [11-19] and project success factors [6-3].
Researchers distinguish objective (like: satisfying the project scope within time/cost/resource
constraints [2] and subjective [19] success factors of the project. In this study objective success factors
that influence the project success are modeled and investigated.
According the Chaos Reports projects can be classified into 3 classes (success, failed and challenging
projects). One of the most interesting results of this report was that the IT projects managed by agile
project management approaches was 3 times more successful than the traditional waterfall projects.
In this study the novel framework of risk analysis system: REACH 4 P 3 (Risk Evaluation, Analysis &
Classification for Hierarchical Process & Project Plans) is used to simulate the changes of the
parameters (like: time, cost and resource demands). There are several novelties in this framework
system: on the one hand this framework system can handle the completion importance of the task
and the probability values of task dependencies as score values; on the other hand new task
completion and new dependencies can also be indicated.

2. Background of the study

In this study matrix-based method is used instead of network-based project planning techniques.
Two kinds of approaches are investigated: agile and traditional project management approaches,
therefore these approaches and these techniques will be compared.

2.1. Agile versus traditional project management

In case of IT projects the application of agile project management approach is often three times
more successful than non-agile like traditional waterfall software developing projects, according to the
2011 CHAOS report from the Standish Group [5] (see Fig. 1).

Figure 1: Success rates of agile and waterfall projects [5]

Comparing the aims and constraints of the project in agile and traditional project planning when
applying agile project planning methods used in IT projects, everything is put up-side-down. While in
traditional construction projects the scope of the project and list of tasks, which have to be realized
are given. The objective function could be the minimal total cost, resource levelling, or minimal project
duration etc. In agile project planning deadlines, resource and cost availabilities are fixed (see Fig. 2).

2
Kosztyán, Zs.T. REACH 4 P3 – Risk Evaluation Analysis & Classification for Hierarchical Process & Project Plans, Global Journal on Technology
[Online]. 2014, 05, pp 01-07. Available from: www.awer-center.org/pitcs

The goal could be the realization of as many tasks as possible; whichever tasks are the most important
for realization.

Figure 2: Comparing the traditional and agile project planning [7]

There are very few methods supporting agile project planning. It is hard to use traditional network
planning methods, but matrix-based methods can handle the specialities of the IT projects [14].

2.2. Network-based versus matrix-based project planning

The most frequently used, traditional project planning methods [8-18-12] methods, Gantt charts
[10], Line of Balance methods [1] primarily support the operative tasks of project planning. Having an
accepted logic plan, where tasks and the dependencies between tasks are determined, we can
schedule tasks and allocate costs and resources by using different kinds of network planning, cost and
resource allocation methods [4]. However, these algorithms slightly support decision-making
problems such as:
 Which tasks or subprojects should be selected for a new project?
 Not having a strict technological order of tasks, which is the best possible sequence of tasks
regarding time, resource and cost demands of the project?
Supporting the logic planning [16] we use matrix-based methods instead of traditional project
planning techniques, aiming to meet management claims.
Network planning methods can also be used for risk analysis and evaluation. Cost/time/resource
demands of the tasks varied, but in case of matrix-based methods the importance of task completion
and the probability of task dependencies can also be changed.

3. Methods
This matrix-based project planning method can handle if a task or a dependency is ignored or a
new/an alternative task or dependency will be specified to the project plan, therefore different kinds
of project scenarios and project plans can be simulated. Some of them will be failed projects, but
others will be successful ones.
Score values can be attached to the task completion. This score value is equal to 1, if the task is
mandatory. The task score value is positive, but lower than one, if the task is supplementary. This
score value can explain the relative importance of task completion. The score value of a project

3
Kosztyán, Zs.T. REACH 4 P3 – Risk Evaluation Analysis & Classification for Hierarchical Process & Project Plans, Global Journal on Technology
[Online]. 2014, 05, pp 01-07. Available from: www.awer-center.org/pitcs

scenario will be the sum of the score value of realized task completion. In this way more task
completions produce higher score value of the given project scenario. The logic plan of a project
scenario is feasible, if every mandatory task decided to be realized and the score value of a project
scenario is not lower than a constraint. A project scenario is feasible, if the logic plan of the project
scenario is feasible, and the time/cost/resource demands of the project are not greater than the
constraints. The scope of the project is characterized as minimal requirement of a project. Minimal
requirement always contains mandatory task completions and rarely any important task completions
which are described as their score values.

3.1. Groups of project scenarios

Similarly to the CHAOS reports 3 kinds of projects can be distinguished.

1. A project is successful; if every project scope is satisfied, within the time/cost/resource
constraints. In other word the given project (scenario) is feasible.
2. A project is a challenged project; if every project scope is satisfied (the logic plan of the
project scenario is feasible). Although this project (scenario) is not feasible, because time,
cost or resource demands are higher than a constraint, these demands are within a
tolerance.
3. A project is failed otherwise.

3.2. Parameters changing

Different kinds and different types of real IT project plans are considered and various kinds of
parameters will be changed (like: time/cost/resource demands, ratio of task completion, probability
values of dependencies etc.) In this model time-, cost and resource demands, task completions
realization of a dependency, new task occurrences can be characterized as probability values. These
probability values can follow different kinds of distributions. In this way score values of projects can be
calculated for different kinds of projects. This model can also be used to simulate shock effects. It
means that not every, but a given ratio of task parameters are changed drastically. It is assumed that a
higher score value of task completion and task dependency proceed higher probability value of
desired realizations. For example, if the score value of task completion is 1 it means that it is a
mandatory function, therefore the probability of desired realization will be 1. Although shock effect
can change not only time/cost and resource demands, but also the probability values of desired
realizations. The final probability values of the realization of task completion or task dependencies will
be calculated after simulating shock effects. Therefore we can simulate that the mandatory tasks will
not be realized.

3.3. Agents for approaches of agile and traditional project managements

Agents are computer programs, based on scheduling, cost minimizing and resource allocation
algorithms. The aim of these programs is to specify a feasible project scenario from a stochastic
project plan. The Traditional Project Management agent (TPMa) can use traditional time/cost trade-
off methods in order to reduce time and/or cost demands [17-9] of the project, and can use resource
allocation and/or resource leveling algorithms for specifying a time and/or resource constraint
resource allocation [4-15] but the logic plan of project is fixed, the project plan will not be
restructured. The Agile Project Management agent (APMa) can ignore supplementary task
completions, and can restructure project if uncertain task dependency is ignored. In this way the logic
plan can be restructured considering the management claims [14-16-13].

4
Kosztyán, Zs.T. REACH 4 P3 – Risk Evaluation Analysis & Classification for Hierarchical Process & Project Plans, Global Journal on Technology
[Online]. 2014, 05, pp 01-07. Available from: www.awer-center.org/pitcs

4. Simulation and empirical results

In this simulation 2 different real software development projects are considered. Based on
interviews the score values of task completion are specified. In these projects the ratios of mandatory
and supplementary functions are different. The original project plans have been revised by corporate
experiments: considering a task dependency between tasks, if there is no technological proceeding,
but the original project plan consists dependency between these two tasks the dependency specified
as uncertain dependency. It means: we can complete these tasks in a sequence according to the
original project plan, but if there are enough resources and the technology can allow the task
completions can be parallelized and the uncertain task dependency can be ignored. Table 1 shows the
percentage of uncertain task completions and task dependencies according to the revised project
plans. These projects contain time, cost and 3 kinds of (manager, programmer, tester) resource
demands.
At the simulation stage 1: after specifying revised project plans, considering the former project
experiments the expected values of time/cost resource demands based on the original project plans,
but in the pessimistic project plans tasks need 30% more time, 20% more cost and 50% more resource
demands. While in the optimistic project plans tasks need 10% less time, 5% less cost and 10% less
resource demands. It is assumed, that the time, cost and resource demands follow the β distribution.
By using the Monte Carlo simulation the uncertainty of project parameters can be simulated
(uncertainty effect).
The next step is to apply the modified or two-step Monte Carlo simulation. In this case the extreme
or shock effects (like system failures) can also be simulated. In this case the first step is to select tasks,
where the parameters would be changed. In this simulation the selection ratio is 5%. The second step
is to change the parameters for the selected tasks. The difference between the one-step and two-step
Monte Carlo simulation is that in the first case every task’s parameters can be changed, but in the
second step only selected task parameters have been changed. At the same time the shock effect is
higher for the selected task than the uncertainty effects. We assumed that the shock effects are
between 50% and 100%.
At the same time in case of software development projects the scope can also be changed during
the project realization, therefore the score values of task completion can also be changed. At the
simulation stage 2, the score value of task completion and dependencies can be changed. If a task
completion is changed from 0 to greater than 0 it means a new task completion will be specified. In
this case the dependencies, cost/resource/time demands are also generated for these new tasks. If a
task completion is changed to 1 it means that this task completion has become a mandatory function.
For simulating the changes of the specification of the project the two-step Monte Carlo simulation is
applied again. Considering the proposals of the corporate experts 20% of score values of the task
completion can be increased maximum 50% and 10% of new task completion can be generated.
Parallel to the simulation stage 2, agents of TPMa and APMa (see previous chapter) are used to
manage project plans in order to satisfy the constraints and management claims. Time, cost and
resource constraints are 80% of maximal values of time, cost and resource demands. The minimal
requirement is to complete all mandatory functions and reach 80% score value of the sum of all score
values of task completions. The TPMa uses time/cost and time/resource trade-off methods in order to
decrease time demands. In this case the crash cost and resource demands will increase 20%, when
time demands can decrease 10%.
After 10 000 simulations the result of success of projects is shown in Table 1.

5
Kosztyán, Zs.T. REACH 4 P3 – Risk Evaluation Analysis & Classification for Hierarchical Process & Project Plans, Global Journal on Technology
[Online]. 2014, 05, pp 01-07. Available from: www.awer-center.org/pitcs

Table 1: Simulation results

T 01 02 03 04 05 06 07 08 09 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 T 01 02 03 04 05 06 07 08 09 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27
Matrix-based 01
02 01
project 03
04
02
03
planning 05
06 04
07 05
08
06
09
07
=Mandatory 10
11 08
=Compulsory 12
13
09
10
task completions/ 14
15 11
task dependencies 16
17
12
18 13
19 14
20
21 15
22 16
23
17
24
25 18
26 19
27
20
28
29 21
30 22
31
32
23
33 24
34 25
35
36
26
37 27

TPMa APMa TPMa APMa

Stages/Agents
Success; Challenged; Failed projects
Uncertainty effect
Stage 1

Shock effect
Changing scores
Stage 2

5. Summary, conclusion and future works

Despite the fact that the agent of APM cannot produce 3 times more successful projects, the
number of success project scenario is significantly higher than in case of using traditional project
management techniques. In future research an adequate combination of traditional and agile project
management techniques should be investigated.
With REACH framework algorithm model time-, cost and resource demands, task completions
realization of a dependency, new task occurrences can be characterized as score values and these
score values can also be changed by simulation technique. As a result probability of success projects
can be calculated for different kinds of projects and later multi-projects. This study does not consider
multi-projects however REACH framework method can also be used with hierarchically organized
multilevel projects. This model can be used both to simulate shock and uncertainty effects.

6
Kosztyán, Zs.T. REACH 4 P3 – Risk Evaluation Analysis & Classification for Hierarchical Process & Project Plans, Global Journal on Technology
[Online]. 2014, 05, pp 01-07. Available from: www.awer-center.org/pitcs

References

[1] Al Sarraj, Z.M., (1990). Formal development of line-of-balance technique, Journal of Construction Engineering
and Management ASCE, 116(4), 689–704.
[2]Atkinson, R., (1999). Project management: cost, time and quality, two best guesses and a phenomenon it's
time to accept other success criteria, International Journal of Project Management, 17, 337-342
[3]Belout, C. G., (2004). Factors influencing project success: the impact of human resource management,
International Journal of Project Management, 22, 1-11
[4]Brucker, P., Drexl, A., Möhring, R.H., Neumann, K., Pesch, E., (1999). Resource-constrained project scheduling:
Notation, classification, models and methods, European Journal of Operational Research, 112(1), 3–41.
[5]Chaos Manifesto 2012, Standish Group
[6]Cookie-Davies, T., (2002). The “real” success factors on projects, International Journal of Project Management,
20, 185-190
[7]Dalcher, D., & Brodie, L., (2007). Successful IT Projects, CengageLearning EMEA.
[8]Eisner H., A. (1962). Generalized Network Approach to the Planning and Scheduling of a Research Project,
Operation Research, 10(1), 115-125.
[9]Feng, C., Liu, L., & Burns, S., (2000). Stochastic Construction Time-Cost Trade-Off Analysis, Journal of
Computing in Civil Engineering, 14(2), 117–126
[10]Gantt, H. L., Work, Wages & Profit, (1974). The Engineering Magazine, New York, 1910; republished as
“Work, Wages and Profits, Easton”, Pennsylvania, Hive Publishing Company.
[11]Gido, J., & Clements, J. P., (2009). Successful Project Management, Cangage Learning.
[12]Kelley, J., & Walker, M. (1959). Critical-Path Planning and Scheduling, Proceedings of the Eastern Joint
Computer Conference.
[13]Kosztyán, Zs. T., (2012). Challanges of the Project Planning Methods in the 21st Century 2012. Problems of
Management in the 21st Century, 2(5), 46-60.
[14]Kosztyán, Zs. T., & Kiss J., (2013). Matrix-Based Methods for Supporting Logic Planning of IT Projects, Lecture
Notes in Electrical Engineering, 151, 311-318.
[15]Kosztyán, Zs. T., Perjés Z., & Bencsik, A., (2008). Resource Allocation and Cost Reduction by Means of
Alternative Solutions, Innovations and Advanced Techniques in Systems, Computing Sciences and
Software Engineering, (ed. Khared Elleithy), Springer, 556-560.
[16]Kosztyán, Zs. T., & Kiss, J., (2010). PEM – “A new matrix method for supporting the logic planning of software
development projects”, 12th International Dependency and Structure Modelling Conference, DSM'10, 22-
23 July 2010, Cambridge, UK, pp. 97-110.
[17]Prabuddha, De, Dunne E.J., B Jay., Ghosh, C.,& Wells E., (1995).The discrete time-cost tradeoff problem
revisited, European Journal of Operation Research, 81(2), 225–238
[18]Roy, B., (1962). Graphes et ordonnancements, Revue Française de recherche opérationnelle, 25(6), 323.
[19]Thomas, G., & Fernandez, W. (2008). Success in IT projects: A matter of definition?, International Journal of
Project Management, 26, 733-742.