The assessment of risks that threaten a project

• Author: Cristina Maria Stoica - Boris Constantin - Marius Alexa
• Position: Professor, Ph.D., 'Petre Andrei' University, Iasi - Lecturer, Ph.D., 'Gh. Asachi' Technical University, Iasi - Assistant, Ph.D.Candidate, 'Al.I.Cuza' University, Iasi
• Pages: 245-261

Cristina Maria Stoica, Boris Constantin, Marius Alexa

245

LESIJ NO. XIX, VOL. 2/2012



THE ASSESSMENT OF RISKS THAT THREATEN A PROJECT

Cristina Maria STOICA∗

Boris CONSTANTIN

Marius ALEXA

Abstract

A project consists of a number of interrelated tasks whose aim is to produce a specific result.

A project risk analysis consists of analyzing schedule, cost risk, quality of the final product etc. A

cost risk analysis consists of looking at the various costs associated with a project, their uncertainties

and any risks or opportunities that may affect these costs. The distributions of cost are added up in a

risk analysis to determine the uncertainty in the total cost of the project. A schedule risk analysis

looks at the time required to complete the various tasks associated with a project, and the

interrelationship between these tasks. In this paper we want to study the various risks associated with

the project. We start this study with the assumption that a project’s cost and duration are linked

together and also cost elements and schedule durations are correlated. The normal uncertainties in

the cost items are modeled by continuous distributions like the Pert or triangular distribution. For

project schedule modeling the most flexible environment is spreadsheet. We are interested in

building blocks that typically make up a schedule risk analysis (also a cost risk analysis) and then

show how these elements are combined to produce a realistic model. In the same time we want

implement software tools for run Monte Carlo simulations on standard project planning

applications.

Keywords: feedback loops, cascading risks, portfolios of risks, sensitivity analysis, Monte

Carlo simulations, critical path analysis, ModelRisk software

Introduction

A project consists of a number of interrelated tasks, whose aim is to produce a certain result.

Typically, a risk analysis on the achievement of a project implies the analysis of the risk regarding

the plan of the project and its cost. In some cases the analysis may also include other aspects, such as

for instance the quality of the final product. Risk may be defined as the possibility of the emergence

of an event that affects the achievement of technical or cost objectives, or of project terms. The risk is

aleatory, unpredictable, may be favorable, yet most of the times is unfavorable and thus, under these

circumstances the analysis and prevention of risks should be an utmost preoccupation for project

managers.

In this paper one intends to identify the risks that may appear during the executio n of a

project, as well as to identify techniques and methodologies of dealing with these risks. Since the risk

is usually tackled statistically and since the statist ical results are better only if the statistical

distributions used are closer to the distribution of real data, one emphasized here the creation of new

statistical distributions that would fulfill this requirement.

With this aim in view the team leaded by the author of this article created a library of

programs, named DistriRisk which is based on the following four methods of creating new





∗ Professor, Ph.D., ”Petre Andrei” University, Iaşi (email: stoicacristinamaria@yahoo.com)

∗∗ Lecturer, Ph.D., ”Gh. Asachi” Technical University, Iaşi (email: borisconstantin@yahoo.com)

∗∗∗ Assistant, Ph.D.Candidate, ”Al.I.Cuza” University, Iaşi

246 Lex ET Scientia. Economics Series

LESIJ NO. XIX, VOL. 2/2012

distributions: the composition method, the „Rejection Sampling” (RS) method, the „Importance

Resampling” (IR) method and the Metropolis – Hastings (MH) algorithm. Using the random

generation of numbers, one also created a program for the generation of beta distribution, applicable

in the Monte Carlo method. All these new distributions are to be included in the pack of programs

ModelRisk developed by the company VOSE Consulting with its European headquarters in Belgium.

This paper also suggests another way to calculate the cost of a project by applying the central

limit theorem. Since the ordinance of a project is a simulation based on a privileged scenario for each

task, we hereby suggest the use of the Monte-Carlo method, which allows the exploitation of several

ordinances, combining different scenarios for the tasks of the project, an d leading to a probabilistic

analysis of certain information, such as the duration of the project or the probab ility for a task to be

difficult to achieve.

In the future we intend to develop a sampling algorithm based on the Latin hypercube

method, which will provide a sampling method that seems to be random but guarantees the

reproduction of the input distribution with greater efficiency that the one provided by the Monte

Carlo method.

Once this methodology is certified, we will try together with the specialists of the company

VOSE Consulting to tackle the quantitative calculation of risk, as well as the possibility to model it,

so as to evaluate as many projects as possible.

The risks of a project

The risks corresponding to the main stages of a project are the following: the analysis of the

needs, preparing the project and the execution of a project.

The analysis of the needs. There are four categories of risks that should be taken into

account before launching a project: - the risk of competition; - the risk of market (the commercial

situation);- commercial risks (the manufacture of the product, the relation cost / quality); -

technological risks.

The factors that may increase the risk could be the following: - the inexistence or the

incomplete previous research in the field of the project; - a need formulated in a wrong way; -

functions or restrictions unspecified by the user; - functions, whose complexity is inadequately

assessed when analyzing need, one also noticing an under-estimate of the level of difficulty that

requires expensive resources; - non-negotiable functions, imposing thus highly restrictive objectives

from a technical point of view or when it comes to price or terms, choosing some functional

performances without being in fact imposed by the needs; - not knowing norms and regulations

imposed on certain products.

Preparing the project. These analyses and conception considerations exert an important

effect in the stage of the execution of the project. In the second stage the following risks may be

mentioned:

- various flaws and constant hesitations in the first version of the project, to which one may

add an incomplete and less competent technical documentation;

- under-estimating the complexity of the methods and conception procedures (programs,

automatization microcontrollers), which leads to too little time spent for the learn ing of working

techniques;

- difficulties in defining and planning the stages mentioned in the program;

- the wrong assessment of the availability and performance of resources used; generally, there

is the tendency to over - estimate performances and to under-dimension costs, being too optimist in

what regards the terms;