Abstract
Effective Project Management is very important for those SMEs that want to compete with other companies in the global market. Those SMEs working with projects could gain competitive advantage using Earned Value Management (EVM) technique. Nevertheless the SMEs cannot afford to spend a lot of time and money in the process of implementing this technique in their companies. It is shown here an accelerated approach that could help in the implementation of the EVM technique in Small Settings. This is achieved integrating the IDEAL (McFeeley, 1996) model cycle, and practices recommended for EVM implementation by the PMI in their standards (PMI, 2008) (PMI, 2005). As recommended by CMMI level 3 (defined process) it has identified phases, activities, roles, inputs and outputs, work products and measurements needed to carry out the implementation. The resulting steps have been simplified taking into consideration the Small Setting constraints. The process has been validated through case studies and surveys in four projects of two very different organizations, measuring different parameters related to process improvement. A defined process (with reduced complexity) and PMO are key factors that help reducing EVM implementation money an time.
Introduction
The objective
“Earned Value Management (EVM) has proven itself to be one of the most effective performance measurement and feedback tools for managing projects” (PMI, 2005). EVM usage at a global level shows a relatively slow, but steady growth, both in the public and private sectors. A higher growth could be expected, given the simplicity of the theoretical base and the benefits it produces. It is important to note that EVM is a technique recommended by the most authorised management practitioners in the world and by prestigious institutions as the PMI, Project Management Institute.
Some of the processes important to improve are those related with Project Management. A Guide to the Project Management Body of Knowledge (PMBOK Guide®) 4th Edition (PMI, 2008) points out 42 processes distributed in a matrix of five process groups and nine knowledge areas. Every process has inputs, tolls and techniques, and outputs. Perhaps the most important tool and technique is EVM (PMI, 2005), Earned Value Management, as it requires to perform a good deal of the best practices recommended. This has been the reason for choosing the implementation of EVM as objective for the experimentation.
Organization of the paper: First there is a definition of some terms used. It continues with a description of the experimentation method used, the case studies and the survey. Finally there is a resume of conclusions. The experience is carried out under the scope of Software Process Improvement for Small Settings in the University Carlos III of Madrid.
The Small Settings picture
In this work it has been adopted the ESI (SEI, 2006) definition for Small Setting. It considers as such:
- small businesses with fewer than 100 people
- small organizations, within a larger organization, with fewer than 50 people
- small projects with fewer than 20 people
Think small first
How Do You Eat An Elephant? One Bite at a Time (Hogan, 2000), or as one International Process Research Consortium - IPRC sponsor put it, “all improvement happens through small groups.”
Both the implementation cost and time of actual SPI CMMI, Project Management processes, or just EVM, are hard items to be handled by most companies in southern Europe, especially when referring to the Small Settings.
Carlos III University of Madrid is deeply involved in Software Process Improvement, working in close collaboration with leading SPI Spanish companies like Zonnect Redes de Ingeniería (www.zonnect.com). The objective of this research is to apply proven methodologies for the small setting, as part of an overall strategy that finally helps also the big ones: Think small first, eat an elephant one bite at a time.
The orientation of Carlos III research in this field is taken from:
- The policy of the EU through the Framework Programs that deeply promotes R&D in SMEs.
- The recommendation and results of the SEI IPRC Project Charter that is promoting the Improvement of Processes in Small Settings (IPSS), planned to start October 1, 2006.
- The best practices recommended by the PMI set of standards.
- The accumulated experience and results of the research group.
This research is oriented to provide approaches, tools, techniques, and guidance for applying methodology and best practices in Small Settings both for Process Improvement and Project Management.
The IDEAL model is being considered a good guidance for all organizations, but the complexity of the steps/activities recommended is a burden for Small Settings that, generally, do not the money, resources and material needed. To any prescribed IDEAL step it has been applied a filter, using expert judgement analysis in order to decide whether the step could be omitted or not in a Small Setting scenario.
Organization of the paper
First there is a definition of terms, with same examples of factors found in projects. It continues with a description of the method used, followed by some facts on the validation method used and the actual state of the research. Finally there is a resume of conclusions.
Concepts and Definitions
Definitions
Some concepts used in this work are:
Acceleration
The effect of reducing the average time and/or cost needed to deploy a practice or process.
SME
We use SME as a synonymous of Small Setting, as defined by the (SEI, 2006), SEI (2002):
- small businesses with fewer than 100 people
- small organizations, within a larger organization, with fewer than 50 people
- small projects with fewer than 20 people
Small setting represents more than 85% of the whole market. This is the reason why the institutions at any level, whether local, national or international, are very interested in giving these small setting the innovation and competences they need to compete in the global market. Here is a justification of our interest in this part of the market.
Experimentation Methodology
As important as the tool and technique we want to implement is the operating subject that is going to use it: a SME. Due to the special characteristics of an SME a special tailoring of the IDEAL model recommendations has been made. The process followed is the following research phases:
- First: a complete IDEAL WBS check list has been made. (Exhibit 1) This WBS is independent of the size of the company were the improvement is going to been made
- Second: This WBS has been augmented including extra activities specially recommended by several authors (Exhibit 2) (PMI, 2005; Cabezas L. et al, 2007a)
- Third: It has been decided whether a particular activity in the WBS was going to be applied or not to a Small Setting. This has been done using “Expert judgement” technique. The chosen activities have defined in writing
- Fourth: Experiment 1. Using “Expert Judgement” it was selected a subset of the Small Setting WBS in order to measure and analyze how they contribute in the acceleration of the implementation process.
- Data and Measurements carried out of four case studies in two very not similar Small Settings.
- Survey
- Analysis and results
- Fifth: Experiment 2: Using “Expert Judgement” it was selected a subset of Experiment 1 WBS in order to compare, using measures and analyses, how a Small setting would perform in two very different scenarios (with and without a special purpose SaaS Platform with ready to use process embedded in it).
- Data and Measurements carried out of four case studies in two very not similar Small Settings, both with and without use of a SaaS platform.
- Survey
- Analysis of results
The research moves along a previously defined roadmap, dealing first with EVM implementation acceleration and second the influence of using a SaaS platform. The model is refined through iteration and experimentation results.
Exhibit 1: A sample of IDEAL based WBS
Exhibit 2: Activities in de WBS directly related to EVM
Case studies, scenarios and survey
Description of the SMEs used in the case studies:
One SME is a very small EBCT (Science and Technological Base Company), with a total staff of 20 people. All have high level of formal education, some are PhD degrees. Nonetheless, around 50% of them have only work experience in the academic world. They are under time pressure for succeeding in two ambitious R&D projects, with fixed deadlines. The SME was constituted one year ago, and most of the staff (Lawyers, Telecommunications, Industrial and Aeronautical engineers, Software Development engineers, Business Administration, Sociology, etc) were hired in the last six months. The end users are very motivated as they constitute what is understood as a “Living Laboratory”
The other SME is the IT Department of an Organization belonging to a Governmental Institution. As such is a Small Setting both because the project involves less than 20 people, and because this IT Department is small organization, within a larger organization, with fewer than 50 people. Nearly all are long standing public employees with an average of 15 years in the company, working from 8:00 to 15:00. They are not accustomed to work under pressure, and these projects are no exception. The subcontracting company carries out the biggest burden of the project, working long hours till 18:00. They work under time and material type of contract. The end users form a “special part” of the project as they are the people that validate the project results and usually are very slow producing the validation reports.
The Case Studies are the four projects. All these projects are monitored and controlled using EVM with regular weekly meetings were the EVM reports supply the main performance information on the project progress. (Exhibit 3)
The survey is constructed with a standard set of questions aimed to asses the project manager and sponsor consideration at project start, middle and end.
Exhibit 3: Summary table of Projects used in as Case Studies
Factors, variables and measurements
For every case study distinction is made between Factors, Variables, Measurements, Comments, and Indexes
For Factor are Environmental parameters, characteristics, constraints, that characterizes a particular project. Some examples: budget, personnel skills, language barriers, time to finish, organization constraints, etc. Factors are catalogued, analyzed and documented (Heales, 2002; Kemerer, 1999; Khosrowpour-pour, 2005). For the sake of the experiments the factors considered are the following (Exhibit 4):
Exhibit 4: Summary of factors considered
Variables, in the context of these experiments, are case study data that changes for whether a SaaS platform is used or not. They are used mainly in Experiment 2. (Exhibit 5)
Exhibit 5: Summary of variables used
Measurement, in the context of these experiments, is the data obtained through objective measurements made during the life of the project. These data are independent of the project method used for monitoring and control, and depend only on case study project performance. (Exhibit 6)
Exhibit 6: Summary of measurements made
Comments are explanatory information that help to understand the Case Study scenario. Indexes are calculated numbers that categorize the results. The indexes are normalized to 1. These indexes are used for experiment 2. (Exhibit 7)
Exhibit 7: Indexes used
Analysis of results
For every case study distinction is made between Factors, Variables, Measurements, Comments, and Indexes. Measurements and analysis has been carried out during the life cycle of the project, and the calculated results have been updated with every measurement made.
First of all, it has been compared Indexes of the four study cases, obtained through surveys on the project manager and sponsor. The results obtained are shown in the following graphic (Exhibit 8).
Exhibit 8: Illustration on Indexes analysis
The indexes have been calculated averaging the PM and Sponsor responses in each case study. It can be observed that the indexes obtain better numbers at the end of the project that at the beginning, fact that was in some way predictable due to the education obtained through the project life.
Conclusions
It has been possible to conclude, according the use of the accelerated Process used, that there is an increase in all the indexes (from start to end surveys). This implies that there is a positive contribution of the accelerated Process for Implementing EVM in all the Case Studies. One curiosity is that the Company (LQT) with a lot of academic knowledge behaves more positively towards the use of the agile process, being the governmental Small Setting (INA) with a lot of civil servants that were more reluctant to use it. The result shown here, are going to be double checked with further surveys.
The conclusions reached during the experimentation can be summarizes as follows:
- It is good to have a Defined Process to accelerate the EVM implementation in an SME, but the formal processes have to go under a high simplification. The defined process is important also because it is a good help having beforehand in writing what has to be done.
- The success depends also of the existence of a PMO, manned at least with a part time person. This helps to consolidate the performed actions and provides continuity with other projects.
- The use of a SaaS platform introduces a further acceleration as it minimizes the time and cost of some of the tasks to be performed, and obliges the continuity of the periodic actions. Nevertheless the effect of the PMO cannot be forgotten. At least an part time PMO should be always present.
