Abstract
In response to the need for professional education on project management (PM), which has brought about a demand for courses with shorter duration, better quality and lower costs, the number of available online educational resources has increased. However, most of existing Distance Learning (DL) environments for project management do not offer individual student assistance, which may hinder the quality of their students' learning. Furthermore, such environments do not deal adequately with the student's lack of motivation and the feeling of isolation, which contribute to high rates of evasion. Artificial Intelligence can be applied to build DL environments that take students skills into account, respect their characteristics as individuals, and thus contribute towards a more effective learning experience. The Project Management Knowledge Learning Environment (or PMK) is a project that aims at building an intelligent learning environment for project management education, and deals with the problems mentioned above. The PMK domain content is based on the PMI's PMBOK and other references on PM. PMK has pedagogical resources as: exercises, PMP mock tests, tips, FAQ, related web links, glossary, notepad, useful templates for the project manager etc. During the project's development, we have also carried out an investigation of the studying strategies used by PMP and by PMP candidates. This enabled us to model teaching strategies in PMK that are adequate and efficient for different types of students. In order to enhance the system's ability to care for individual students, an intelligent pedagogical agent called VICTOR (Virtual Intelligent Companion for TutOring and Reflection) was built. He interacts with the students and cooperates with their learning process. VICTOR, an animated character, behaves as a chatterbot (which answers questions about the domain). He provides immediate feedback for student's actions, gives tips and tries to keep the student motivated. PMK is an open-source software project, implemented as a web site, which aims to disseminate PM knowledge and promotes intelligent and attractive learning experience. This paper presents the contribution of the PMK Project for PM education.
Introduction
To take full advantage of the benefits of project management (PM), organizations must be aware not only in adopting project management as a profession, but as a methodology in which project managers should be trained, as a way to add value to their individual experiences. PM should be treated in a professional manner and conducted by qualified people. In this way, an organizational project management culture should be created and its adoption should be implemented in a systematic way, so its principles can be practiced satisfying the organizational needs.
Niskier and Blois mention that successful professionals need to be ready to recycle their knowledge and add concepts, behavior and attitudes (2003). Continuing education is a possible way to reach this objective. Senge states that organizations learn through individuals that learn (1990). The individual learning does not guarantee the organizational learning, but without it the organizational learning does not occur. The fundamental competence to assure enterprise continuing and flourish is the ability to learn. The professional education reflects on the success of the organization.
Distance learning (DL) is an alternative for the needs of education and training of professionals and organizations. According to Dalmau, Rodrigues, Valente, and Barcia, DL is considered a methodology oriented to adults with professional and familiar commitments, because it allows the continuation of the study without hindering other activities (2001). However, various DL educational systems do not delivery because they do not treat adequately the motivation problem of their users and because a very impersonal presentation is used, which in turn influences the student performance (Jonhson, Rickel & Lester, 2000). In the majority of those systems, there is not a constant physical contact with colleagues and teachers, as one can find in a traditional classroom. This problem is even worse when students find themselves in trouble to overcome, very often, quite simple learning barriers and have no one to aid them. In this environment, the student perceives an isolation feeling, which leads to many drop outs in traditional web educational systems (Hara & Kling, 1999).
Valente mention that learning involves the constant accompaniment of the apprentice, in order to understand who s/he is and what s/he does, to be capable to propose challenges and help her/him to assign meaning to what s/he is performing (Valente, 2003). Only in this way, the apprentice is able to process information building new knowledge. The biggest problem of DL traditional systems is a generic approach towards learning (i.e., incapability to treat individually the skills, knowledge and performance of the students). Frasson believes that “various institutions are using e-learning as a set of tools where everyone can talk to each other, as a generalized e-mail system, and this is completely inefficient in terms of good learning” (2003).
Considering those observations, the use of a virtual Learning Companion (LC), can help to solve or minimize the problems just mentioned, influencing positively the student's performance (Chou, Chan & Linc, 2003). The use of LC in DL systems characterizes i-learning (intelligent learning) which objective is to treat the problems found in e-learning and provide individual solutions which take into account the personality and skill of each student, respecting his/her individuality and characteristics, and contributing towards a more effective learning (Frasson, 2003).
This article presents the Project Management Knowledge Learning Environment (PMK), an intelligent learning environment (ILE), that teaches project management methods and techniques. This environment provides a general overview on what is needed to be a project manager and uses a LC as a support technology, VICTOR (Virtual Intelligent Companion for TutOring and Reflection). This LC collaborates with the student learning providing immediate feedback to the student's actions, besides tricks and incentives to study. The PMK aims to treat the evasion problem found in traditional DL systems, offering an individual and customized education.
The PMK Learning Environment
Motivation
Professional development within a short time duration, with competence, quality and reduced costs in order to manage successful projects, in the private and public sector, arises as a consequence of the actual project environment. Organizations which are in a global and increased competitive environment, with quick and big changes need innovation in its products and services. Project managers should be prepared professionals to perform well its roles bringing desirable benefits to their organization. Efficient project management in a time of great change is one of the greatest challenges of moderns executives (Kerzner, 2001).
A way to overcome this challenge is the development of skills and knowledge to reach a professionalism level compatible with the correspondent function. The PM professional needs to learn basic concepts, techniques and management tools, as well as its practice (Project Management Institute, 2000).
However, there is a shortage of resources for PM training and development (team leaders), mainly in the public sector, where resources are scarce. The PMK objective is to training and develop PM professional or people interested in PM, using the web, in an accessible way, with quality, reduced costs and optimizing the professional available time. PMK is an environment that also can be used for PM continuing education.
Features
The PMK study content is based in the PMBOK Guide (Project Management Institute, 2000), which provides basic knowledge and practices on project management, and other bibliographical references on PM (Kerzner, 2001; Mulcahy, 2002).
The PMK has pedagogical resources (functionalities) which enrich the learning of PM: multiple-alternatives exercises (to test knowledge), exercises to assemble processes diagrams (to memorize process items), fill in the gaps exercises (to review and learn concepts), PMP mock tests, student performance reports (finished exercises, difficulties, accessed contents), FAQ, PM library (including files on PM), tricks, e-mail questions and concepts, internet links, glossary, note pad, list of figures, templates, among others. The main objective of the PMK tool is to offer to the student a more practical and interactive view on project management, besides a good preparation tool for the PMP certification exam.
The PMK tool has a LC, called VICTOR, an intelligent pedagogical agent that collaborates to the student learning process. The use of VICTOR in the PMK tool helps to observe, stimulate, guide the learning and gives suggestions to the student (Johnson, Rickel & Lester, 2000).
VICTOR has an animated character (Correia 2004). The use of an animated character can improve the communication with the student, increases his/her motivation and encourages him/her to think deeply about his/her progress (Johnson et al., 2000). It is worth to mention that the LC must interact with the student, providing immediate feedback, without distracting the learner, mainly when the student is executing an improper action (Qu, Wang & Johnson, 2004). VICTOR also behaves as a chatterbot (Leitão 2004) answering some questions about the PM domain.
Research Methodology
PMK is a multidisciplinary project that aims at investigating techniques for designing and developing ILEs. This research involves the combination of knowledge from various disciplines, such as Project Management, Artificial Intelligence in Education, Psychology and Software Engineering. The methodology adopted for this project's execution consists of the steps described in the following sections.
Acquiring and Maintaining Domain Knowledge
When developing an environment that teaches a given domain, the first step is to get to know the domain as well as possible. Thus, we have performed a thorough study of the PMBOK as well as of other relevant references in the area (e.g Kerzner, 2001; Mulcahy 2002). As a result, the PMK's knowledge domain is based on the contents of PMBOK.
In the sequel, we built an ontology that formalizes the PMBOK content. This is done through the representation of concepts, terms and their relationships (represented by predicates and functions) (Gruber 1993).
The ontology model for PMBOK, showed in Exhibit 1, allows us to visualize the relationship between domain terms (e.g. the WBS belongs to inputs and outputs of different processes). This domain representation enables VICTOR to identify, for example, which processes use the term WBS as well as the learning difficulties associated with this term (via errors committed by the learner in exercises that relate to WBS, or via checking whether the student has accessed related terms). The ontology is also used by the chatterbot for answering questions about the domain. Examples of such questions include (Leitão 2004) “what are the entries for the Schedule Control process?” and “To which process group does the process Schedule Control belong to ?”.
Once we had gathered and formalized the domain knowledge, the next step was to identify the main learning difficulties in PM.
Identifying the Main Learning Difficulties in the PM Domain
A member to the research team enrolled in a face-to-face course that prepares students that wish to take the PMI certificate exam. This course allowed the researcher to observe the main learning difficulties related to the PMBOK content, as well as the teaching strategies and hints used by the instructors.
In order to deep our understanding of the learning difficulties related to the domain we have put a questionnaire up on the web, asking both PMPs and PMP candidates details about their studying strategies. This study counted with 147 participants. Among these, 21.77% already have the PMP certificate, and 61.25% have between 2 and 6 years of experience in PM. Other interesting findings are detailed below.
97.28 % of the participants had used PMBOK to study for the exam. In order to get a better understanding of the possible learning difficulties, we asked participants to classify (from a list of 20 items) the five tools and techniques they considered most difficult to learn. The Earned Value Management (EVM) came out first, with 24.49%; in second we had Earned Value (EV), with 13.61% of the votes; Scope Change Control System came in third with 6.80% of the votes, Make-or-Buy Analysis came in fourth with 4.76% and finally there was a draw among Cost Change Control System, Communications Skills e Contract Type Selection with 4.08%. The study also shows that 85.00% of the participants are motivated to learn PM and to use the Web to study for the PMP certificate exam.
Exhibit 1 – Domain Ontology of PMBOK
According to the respondents the most difficult knowledge area is Risk Management. This finding can be related to the fact that organizations do not have a standard process for this managing risks. Instead they only keep track of the best practices used. Risk management activities are used to the existing processes in each organization, as pointed out by Gusmão and Moura (2004).
On the other hand, participants have considered Scope Management to be the easiest knowledge area. This can be related to the daily practice of professionals that deal with projects, and, with or without a methodology, and the need to define a minimum scope for their project, product or service. Our study has also pointed to other relevant aspects, such as books, software tools and tips used when people study for the exam (cf. Torreão 2004).
Usability and Comparative Analysis
In parallel with the identification of possible learning difficulties we have carried out a usability analysis and competitors' analysis in order to help us define the PMK scope.
The usability analysis (Pinheiro, 2003) performed for the PMBOK Easy Project (Rodrigues, 2003) was the starting point for the design of the PMK's Graphic User Interface (GUI). Our project followed Nielsen and Tahir's recommendations (2002). We have also carried out the comparative analysis of strengths and weaknesses of the functionalities of web sites and software tools similar to the PMK tool (c.f. Santos, Gomes & Torreão, 2004). The comparison criteria were exercises provided, domain contents and other resources (such as glossary and tips). As a result, we can list as PMK's strengths the availability of multiple choice exercises for both study mode and PMP mock tests; as weaknesses, we can list the exhibition of tips when the student is solving multiple-choice exercises in the study mode. Once these analyses were finished, the PMK's scope was defined, and its implementation has began.
Definition and Implementation of the System's Functionalities
After the usability and comparative analyses were done, PMK's functionalities and GUI were defined and implemented. The PMK's GUI was designed in a way that allows the student to choose the way s/he wishes to study project management: by process groups, by knowledge areas or by processes. PMK has a set of functionalities, which were derived from the strong points observed in the comparative analysis, together with other functionalities such as selecting exercises using a filter that match a word (like Earned Value). Once that step was concluded, the Learning Companion was added to PMK.
LC's Design, Implementation and Integration to PMK
After the construction of the PMK's GUI and the implementation of its functionalities, the LC VICTOR was built and integrated to the PMK. VICTOR's construction was based on the methodology proposed in (Torreão, Aquino, Tedesco, Sá & Correia, 2004), which presents 6 stages.
The two first stages of the methodology for the construction of the LC (namely, Identifying the Problem and Eliciting Relevant Domain Concepts), were described in the item Acquiring and Maintaining Domain Knowledge of this article. In the third stage, Conceptualizing the Pedagogical Task, the objective of VICTOR and the pedagogical strategy and tactics used by him were defined. VICTOR's objective is to collaborate with the student, promoting learning. VICTOR uses different pedagogical strategies (e.g motivational) with different tactics to fulfill his objective.
In the fourth stage, Building the LC's Architecture, VICTOR's behavior was modeled according to the defined tactics, strategies, and goals. VICTOR's behavior determines how, when and what he perceives from his interaction with the student and environment, as well as what are his answers to the student, (which take into account information stored about the student, such as abilities, knowledge, performance).
For each of the student's action, VICTOR applies different tactics. For example, he may supply immediate feedback, praise the student for his/her effort, praise the correctness of tasks, point errors made, and motivate the student to persevere. Depending on the situation, VICTOR may also point learning difficulties out, give tips to the student, encourage the student to reflect on his/her progress, and answer to some of the student's questions.
The Learning Style versus Personality Type approach is dealt with by the PMK during the application of pedagogical tactics (Durling, Cross & Johnson, 1996). When registering with the system, the student can fill a test that evaluates his/her type of personality, based on the Myers Briggs Type Indicator (MBTI® test) (Myers & Briggs Foundation 2002) and VICTOR to modify the pedagogical tactics. If the student does not fill this test, VICTOR uses his standard tactics.
Still in this stage of the methodology, the Domain Knowledge Base was constructed. To make things easier for the Knowledge Representation task, a XML diagram of the ontology was built (Exhibit 1). The architecture of the implemented LC had four main components: Student's Model that stores the individual information about each student and provides information to the pedagogical module; Pedagogical Module that provides model of the learning process and contains the model of behavior of LC; Domain Knowledge module that contains information about what you teach and a Communication Module that mediates the LC's interaction with the environment.
At the end of this stage, the character's animations (Correia, 2004), used to represent VICTOR's emotional states (e.g. surprise, when student makes mistake; and joy (Exhibit 2), when the student answers correctly to the exercise) were defined. A chatterbot (Leitão, 2004) that answers some questions on the domain was integrated to VICTOR.
Exhibit 2 – Frame showing VICTOR's gestures when the student does an exercise correctly
In the fifth stage, Implementing the LC, the architecture components and animations were implemented. A prototype of VICTOR was integrated to the PMK. After concluding the integration, the last stage of the methodology, Assessing and Refining the LC, was carried out. At this point other functionalities (e.g. generation of progress reports) were implemented and tested. The prototype of VICTOR was validated and adjustments have been made in order to transform it into a full running version.
Conclusions and Future Works
Efficient project management is not only a big challenge but also a critical factor for the success and survival of organizations. This challenge requires an effort to make those organizations aware of the needs and subsequent benefits of adopting project management methodologies and training theirs teams and project managers. In order to tackle this situation, we have developed the PMK, an integrated learning environment that can be used to educate professionals, students and people interested in managing projects in a more effective and low-cost way.
The preliminary experiments carried out with the PMK environment have yielded promising results related to the evidence of the greater efficiency of an intelligent learning environment. This type of system, specially those which count on LCs, help to overcome the main difficulties of the Distance Learning systems: discouragement and dropouts. The LC here proposed aims at meeting the student's needs in a motivational, dynamic and intelligent way.
In the future, VICTOR will also be able to use the Case-based Teaching Strategy as a pedagogical strategy, presenting PM real-world project scenarios. VICTOR may be extended to simulate pair-studying of students. Adaptive tests which stimulate the student self-confidence could be incorporated to the multiple-choice exercises. Another researcher in our group is working on improving VICTOR's interaction through the use of natural language.
In the near future, we intend to carry out more comprehensive tests with users aspiring to PMP certificates and undergraduate students from Informatics Center, Federal University of Pernambuco. by comparing the performance of those who used LC with those who did not.
The PMK system is an open source software research project funded by the Brazilian Research Council – CNPq. For additional and detailed information about the PMK Project access the project website at www.cin.ufpe.br/~pmk.
“PMBOK, PMI and PMP” are trademarks, service marks or certification marks of the Project Management Institute, Inc., which is registered in the United States and other nations.