Project knowledge management: An ontological view

In this research, “Domain Ontology for Project Knowledge Management” is presented by literature and reliable resource reviews and analysis in three layers: “People”, “Technology” and “Process”. This ontology consists of 115 cells. The layer of “People” has been divided into two subgroups: “Culture” and “Leadership”, in12 cells. The layer of “Technology” has been classified into two subgroups: “Technology Component” and “Application”, which has 72 cells. Finally the layer of “Process” has been divided into five groups: “Initiating a Project”, “Planning a Project”, “Executing a Project”, “Monitoring and Controlling a Project” and “Closing a Project”, and has 31 cells. Consequently, the proposed ontology has been evaluated by survey research benefiting from experts’ opinions. In this step, by purposeful sampling and the snowball technique, experts in project management and knowledge management scopes have been determined. Using an online questionnaire; the “Domain” of the designed ontology has been evaluated. After confirming the ontology’s domain, the “Quality” of the ontology has been evaluated with the aid of some criteria extracted from literature reviews by another online questionnaire. Accepted by a certainty of Knowledge Management & E-Learning, 8(2), 292–316 293 95% and Friedman Test, the proposed ontology shows that its three layers are homogenous with a certainty of 95% based on statistical analyses.


Introduction
Economic development is characterized by a continuous de-materialization of the value chain. This leads to a growing knowledge-intensity of work contents and the more influencing role of services. As a result, knowledge plays an important role as the intangible resource and asset of organizations (Nahapiet & Goshal, 1998;Teece, 1998). This trend is mirrored by theoretical approaches underlying the relevance of knowledge. The knowledge-based view of a firm considers knowledge and the ability to integrate individual knowledge for a common task fulfillment essential for competitive advantages (Grant, 1996). At the same time, the degree of temporary forms of co-operation and working constellations is growing. The prevalence of projects as a form of organizing has only recently been acknowledged (Saito, Umemoto, & Ikeda, 2007). Nevertheless, many project-based businesses lack the expertise to handle their knowledge assets (Ajmal, Helo, & Kekäle, 2010) or these cases are still equivocal (Chang, Hung, Yen, & Tseng, 2009).
The temporality and uniqueness in a project are the main barriers for organizational learning. This holds particularly true for projects lacking an organizational memory, routines and other mechanisms of organizational learning (Brusoni, Prencipe, & Salter, 1998;Hanisch, Lindner, Mueller, & Wald, 2009). The management of knowledge in and of temporary organizations is therefore an increasingly important and even a decisively competitive factor (Hanisch, Lindner, Mueller, & Wald, 2009). To operate effectively in a dynamic business environment, firms need to ''have a holistic overview of their project knowledge'', their capabilities, and environment. To access this kind of view to project knowledge management, this research has provided "domain ontology". Broadly defined, ontology consists of terms, their definitions, and descriptions of their relationships. Among many other possible benefits, ontology can be used to facilitate common understanding and the sharing of knowledge in a particular domain (Saito, Umemoto, & Ikeda, 2007).
In both research areas of knowledge management and project management, a substantial quantity of theoretical, conceptual and empirical studies have dealt with different questions about respective disciplines. However, little research has been conducted to include both areas (Love, Fong, & Irani, 2005;Brookes, Morton, Dainty, & Burns, 2006;Hanisch, Lindner, Mueller, & Wald, 2009) and there is no study to present a domain ontology for knowledge management in temporary organizations. Thereupon, in this paper, the author presents the domain ontology to facilitate the implementation of knowledge management in project-based organizations.

Project knowledge management
Project Knowledge Management (PKM) is the knowledge management in project situations and thus the link between the principles of knowledge management and project management (Hanisch, Lindner, Mueller, & Wald, 2009).
On a more general level, not only is the knowledge within projects part of PKM, but also the knowledge between different projects and about projects is considered part of it (Schindler, 2002). The knowledge within projects is closely linked to the project management methodology and the communication practices in projects; both are strongly dependent on the project manager and the individual project management style (Hanisch, Lindner, Mueller, & Wald, 2009).
The particular challenges of PKM are caused by the inherent project characteristics (Love, Fong, & Irani, 2005;Schindler & Eppler, 2003). Projects are unique and temporary undertakings with changing work-force. Moreover, projects are often short-term oriented and integrate the internal and external knowledge of experts. Project participants have to adapt quickly to new conditions and contents of work. The temporality and uniqueness in projects are the main barriers for organizational learning. This is particularly true for projects lacking an organizational memory, routines and other mechanisms of organizational learning (Brusoni, Prencipe, & Salter, 1998;Hanisch, Lindner, Mueller, & Wald, 2009). This factor demonstrates the important role of implementing knowledge management in projects. In recent years, project knowledge management has been ingratiated. Some of the related researches have been presented in Table 1.Nevertheless, as it can be seen in this table, most research works are about the best practices, benchmarking, process reorganization, etc. and there is no study about ontological views to project knowledge management.

Ontology
Ontology is a discipline of philosophy that studies different categories of things that exist or may exist in a given domain. The term was borrowed by computer scientists in the mid-1980s as a means to represent information and knowledge. It gained momentum in the 1990s, when it became widely accepted that information systems should be made interoperable (Welty, 2003). A further thrust came with the proposal of the semantic web, an initiative to embed meaning into web pages so that they become machineunderstandable (Berners-Lee, 2000). Current uses of ontology include the development of information systems, application integration, the organization of content in web sites, the categorization of products in e-commerce, structured and comparative searches of digital content; standard vocabularies in expert domains and product configuration in manufacturing among many others (McGuinness, 2002). Ontology can be designed with increasing levels of formality, from simple glossaries and thesauri to rigorously formalize logical theories and the higher degree of formality, the less ambiguity and the stronger power for automated reasoning (McGuinness, 2002;Uschold & Gruninger, 2004). Thereupon, an ontology-based method for knowledge representation offers a means for the reuse and sharing of knowledge unambiguously (Yang, Miao, Wu, & Zhou, 2009 Building trust in inter-organizational projects by focusing on the impact of project staffing and project rewards on the formation of trust, knowledge acquisition and product innovation. (Maurer, 2010) 2 Introducing knowledge management to improve project communication and implementation. (Koskinen, 2004) 3 Providing a detailed review of IT system which is useful for KM activities in variety project contexts. (Leseure & Brookes, 2004) 4 Providing a framework for social processes, patterns and practices and project knowledge management. (Bresnen, Edelman, Newell, Scarbrough, & Swan, 2003) 5 Focusing on knowledge creation in multidisciplinary project teams. (Fong, 2003;Leseure & Brookes, 2004) 6 Process Post-project reviewing as a key project management competence. (Anbariai, Carayannis, & Voetsch, 2008) 7 Enabling knowledge creation and sharing in transnational projects. (Adenfelt & Lagerström, 2006) 8 Technology Focusing on the use of object oriented technology in project based organizations. (Weiser & Morrison, 1998) 9 Process, Technology Constructing a relevant data structure in Project based organizations. (Matta, Ribiere, Corby, Lewkowicz, & Zacklad, 2000) 10 People, Process, Technology Benchmarking of knowledge management in project based organizations. (Hanisch, Lindner, Mueller, & Wald, 2009) 11 Exploring the knowledge inventory in project-based organizations.
( Van Donk & Riezebos, 2005) 12 Presenting a structural model (present three layers for knowledge of project) for knowledge of project based organization: infrastructure, info structure and info culture. (Leseure & Brookes, 2004) 13 Providing a comprehensive discussion of the KM problems faced by IT project organizations. (Disterer, 2002) 14 Reviewing of knowledge management activities in the engineering to order capital goods in project based organizations. (Braiden & Hicks, 2000) 15 Focusing on significance of relationship between PM and KM. (Gilbert & Holder, 2000;Kamara, Leseure, Carillo, & Anumba, 2000) 16 Introducing the COLA review process as an example of a system able to trigger reflection and formulation of lessons learned. (Orange, Cushman, & Burke, 1999) There are many methods for developing ontology, and each has strengths and weaknesses (Chen, Chen, & Chu, 2009). For example, Noy and McGuinness (2001) suggested a process including the following steps: Step 1: determining the domain and scope of the ontology; Step 2: considering the use of existing ontology; Step 3: listing important terms; Step 4: defining classes and their hierarchy; Step 5: defining properties of classes; Step 6: defining restrictions on properties; Step 7: listing examples in classes.
Knowledge in ontology is the formalized application of five kinds of components: concepts, relations, attributes, axioms and instances (Gruber, 1993;Gómez-Pérez & Benjamins, 1999;Studer, Benjamins, & Fensel, 1998):  Concepts are used in a broad sense. A concept can be anything about which something is said and therefore, could also be the description of a task, function, action, strategy, reasoning process, etc.  Relations represent a type of interaction between the concepts of the domain.  Attributes are functions and attributes of concepts.  Axioms are used to model sentences that are always true.  Instances are used to represent elements.
Once the main components of ontology have been represented, the ontology can be implemented in various languages: highly informal, semi-informal, semi-formal and rigorously formal languages (Uschold, 1996).
One of the most important steps in designing ontology is "ontology evaluation". There are several researches on ontology evaluation, which are briefly expressed in Table  2. In order to assess the accuracy and appropriateness of ontology; its domain must be evaluated (e.g., whether the proposed subgroups are in the determined domain? Whether these subgroups cover the whole headers? …) followed by the analysis of the quality of covering based on the acceptance of domain covering, (Gómez-Pérez & Benjamins, 1999). Some criteria for this type of evaluation are presented in table 2. Based on these criteria, the evaluation methodology has been determined in section 3.

Methodology
This research consists of two basic steps. Firstly, the data were collected from literature and other reliable review sources to be analyzed. The most important concepts in project knowledge management were determined; then with regard to their functions, the domain ontology for knowledge management, consisting of "Concepts", "Attributes" and "Relations" was presented. Comparison the ontology with a reference model for evaluating the ontology producing process (Yu, Thom, & Tam, 2007) 2. Criteria-Based Approach Comparing the ontology based on some criteria and appointment the credit to every ontology for comparison by experts' opinion (Brewster, Alani, Dasmahapatra, & Wilks, 2004;Yu, Thom, & Tam, 2007)

Task-Application-Based Approach
Comparing several ontology in same scope from a specific task point of view (Porzel & Malaka, 2004)

Data Driven Approach
Comparing the ontology based on the data recourse that used for producing the ontology (Porzel & Malaka, 2004) Quality Criteria For Ontology Evaluation In the second step; the proposed ontology was evaluated with respect to "domain" and "quality". The process of quality evaluation was followed by "after modeling evaluation" approach, "criteria-based approach" and beneficially "clarity", "compression", "accuracy", "universality", "expansion" and "stability" quality criteria in "lexical, vocabulary, or data level" and with the aid of "accuracy", "universality", "expansion" and "stability" quality criteria in "hierarchy or taxonomy level" as well as "semantic relation level". Furthermore, "confirming the ontology by expert society" ( i.e. knowledge management and project management experts) solution was utilized for this evaluation. The evaluation process is extracted from Table 2.
In this step, the ontology was evaluated by survey research beneficially of experts' opinion. Initially, by purposeful sampling and the snowball technique, experts in project management and knowledge management scopes were determined. Then through an online questionnaire, the "domain" of the designed ontology was evaluated. After confirming the ontology; the "quality" of the confirmed ontology was assessed by using some criteria derived from literature review by online questionnaire.
The "Domain evaluation" questionnaire contained 75 questions and the "quality evaluation" questionnaire involved 42 based on Likert scale. Some open questions were added to both questionnaires to include other points of view.
Based on statistical analyses (Binomial and Mean tests), the proposed ontology was tested. There with, by Friedman test, the equality of three layers of ontology was examined. The examined hypotheses are:

 Domain evaluation:
 Hypothesis 1: Experts 'opinions in the first questionnaire will follow the normal distribution.  Hypothesis 2: The domain of the ontology is confirmed by experts.  Hypothesis 3: The three layers of the ontology are homogeneous (from "domain" point of view).  Quality evaluation:  Hypothesis 4: Experts' opinions in the second questionnaire will follow the normal distribution.  Hypothesis 5: The quality of the ontology is confirmed by experts.  Hypothesis 6: The three layers of the ontology are homogeneous (from "quality" point of view).

Step one-ontology design
As mentioned before, in this research the "domain ontology for project knowledge management" has been presented by literature and reliable review sources and analyses in three layers of: "People", "Process" and "Technology". "People" has been divided into two subgroups: "Culture" and "Leadership". "Technology" has been classified into two subgroups of: "Technology Component" and "Application". "The layer of Process" has been divided into five groups: "Initiating Project", "Planning Project", "Executing Project", "Monitoring and Controlling Project" and "Closing Project".

People
The category of "People" can be divided into two subgroups: "Leadership" and "Culture". In project-based organizations, the stream of knowledge culture in all areas of organization and projects life cycle is evident. On the other hand, organization culture is influenced by organization leaders and their power that can influence values, attitudes and beliefs. Hence selecting the preferred culture and leadership style based on project knowledge management strategy is extremely important for the successful implementation knowledge management in projects.
In terms of the culture and leadership of these organizations, human resource management with a knowledge approach is the most important factor for training and persuading people by establishing compatible a "performance evaluation system", "payroll system", "pension system" etc., for individuals, groups and the entire organization, which can increase trust (Maurer, 2010) in sharing and applying knowledge in projects. In such confident environments, trust, belief and finally the knowledge-based culture will be thematic in projects and the people of organization can align other strategies with knowledge strategies. This strategy alignment can integrate other layers, such as "Technology" and "Process" with "People". In Fig. 1, "People" can be seen as a layer of domain ontology for project knowledge management. Culture: The importance of culture in project knowledge management has been extracted from literature review. Thereupon, this significance has been rendered a "culture" as a substratum in proposal domain ontology. Cases with specific cultural concepts of knowledge management project are described in Table 3. Cultural concepts are divided into four groups: strategic awareness, collaboration, trust, and keeping current culture.  (Hanisch, Lindner, Mueller, & Wald, 2009) 2. Permanently secure the knowledge gained during projects is the establishment of reward systems for enhancing the security of expert information and therefore create trust. 3. Particularly openness, transparency, the prioritization of PKM related activities and the dealing with mistakes Keeping Current Culture 1. "Keeping current culture"; by use of newsletter, workshops and training. (Leseure & Brookes, 2004) Leadership: The significance of leadership in project knowledge management, extracted from literature review has made "leadership" a substratum in proposal domain ontology. Cases with specific leadership concepts of knowledge management project are described in   (Saito, Umemoto, & Ikeda, 2007).
In this research business-driven one translates to project based applications.
There are various studies on KM process; emphasizing the importance of processcentred knowledge approach (Han & Park, 2009). Notwithstanding the quantity and variety of them, four building blocks in KM process are common. These four basic KM processes are: "Create and Capture Knowledge", "Coding and Storing Knowledge", "Distribution and sharing Knowledge" and "Learning and Applying Knowledge".
Furthermore, the understanding of KM technologies in terms of knowledge processes can be misleading, since those processes are heavily context-related and subjectively interpreted. Hence expressing them in terms of the four types of support to functions uncovered in the review of KM strategy and KM processes has been suggested (Saito, Umemoto, & Ikeda, 2007):  Collaboration technologies: supporting the creation of knowledge according to a personalization approach.
 Dissemination technologies: supporting the transfer of knowledge according to a personalization approach.
 Discovery technologies: supporting the creation of knowledge according to a codification approach.
 Repository technologies: supporting the transfer of knowledge according to a codification approach.

Fig. 2. The "Technology" layer in the domain ontology for project knowledge management
Based on these four groups, in Fig. 2, "Technology" has been shown as a layer of domain ontology for project knowledge management.

Technology component
A comprehensive survey of technologies is a challenging task since their quantity and variety are astounding. Their integration in multiple levels even compounds the task. Here, a fairly extensive list of component technologies is presented, which is classified according to functionality to facilitate understanding (Saito, Umemoto, & Ikeda, 2007):  Storage: Databases, repositories, file-servers, data warehouses, data marts, etc.
 Connectivity: Internet, security, authentication, wireless networking, mobile computing, peer-to-peer, etc. These myriad technologies can support KM in multiple ways, fitting more than one of the collaboration-dissemination-discovery-repository categories. Fig. 3 demonstrates the functional classification according to their most relevant types of support to functions (Saito, Umemoto, & Ikeda, 2007).  Content management: Manage the whole Web publishing process. Manage authors and the content creation process, separate content from layout for standardized output, support multimedia repositories, automatic page-generation via templates, and staging of new content.  Process management: Also known as workflow, automate the flow of tasks and information across business processes. Include workflow engines for handling cases, and tools for modeling processes, accessing external applications, and monitoring and managing operations.  Group support: Also known as groupware, support the work of groups and teams.
Include tools for communication, coordination and collaboration.  Project management: Support the management of project activities and resources.
Include functions for defining and organizing activities and tasks, assigning responsibilities and deadlines, allocating personnel and other resources, and identifying milestones, critical paths and constraints.  Community support: Coordinate interaction in large groups. Include tools for communication and interaction, management of participation levels, including leading and facilitating roles, identity profiling, and collective decision making.  Decision support: Also known as business intelligence, integrate a series of tools for decision making. Include query and report of operational data, managerial dashboards like the balanced scorecard, and decision models and techniques for structured and unstructured situations.
 Discovery and data mining: Support the identification of patterns and associations in large amounts of data, including tools for cleaning and organizing data into data warehouses, and a series of analytical techniques and visualization tools.
 Search and organization: Facilitate access to and organize unstructured content.
Identify key words and topics in documents from varied sources, generate indexes and taxonomies automatically, categorize documents in topics according to relevance, and use domain-specific ontology for specialized classification. Include functions like identification and profiling of experts, communication tools for questioning and answering, rating of answers and experts, and repositories for reusing contributions.
Although each type of KM application has some functionality to fit other quadrants, the main purpose and core function of the application best suits one of them. Fig. 4 represents the functional classification according to their most relevant types of support to functions (Saito, Umemoto, & Ikeda, 2007).   Collaboration and Conferencing: If users are working with a remote team on a project, they are probably going to need some online space to collaborate and meet, whether it is supposed to work on general concepts or to work out specific bugs. Here are some solutions to help users collaborate with those on their team or with their clients. Some of these applications are: ActiveCollab, DinDim, Vyew, etc.
 Invoicing: Unless users are working on an internal project, chances are they will need to send out invoices. Have an invoice program that also makes proposals is vital, as is having one that integrates directly with project management application. Some of these applications are: Simply Invoices, Less Accounting, etc.
 Time Tracking: Whether users need to keep track of time for billing purposes, for their boss, or just to measure their own productivity, chances are they will need a time-tracking application. Some of these applications are:LiveTimer, fourteenDayz, etc.
Although each type of PM application has some functionality that fits other quadrants, the main purpose and core function of the application best suits one of them. Fig. 5 represents the functional classification according to their most relevant types of support to functions. Special project management applications are used in projects based on the type of projects, such as constructional, IT, R&D, etc... For example, the software used for product design belongs to this group and these applications can fit into four categories of "collaboration-dissemination-discovery-repository", based on their functionality and nature.

Process
Four basic processes can be defined for knowledge management: "Creating and Capturing Knowledge", "Coding and Storing Knowledge", "Distribution and sharing Knowledge" and "Learning and applying Knowledge". On the other hand, project management processes can be defined in five phases. According to knowledge layers in project management, in order to conflate these two types of processes (project management and knowledge management), two building blocks "Setting Knowledge Goals" and "Knowledge Evaluation" based on Probst model (2002) were added to knowledge processes.
In Fig. 6, "Process" can be seen as a layer of domain ontology for project knowledge management.

Initiating a project
Initiating a project is the first phase of projects. The integration of "knowledge management processes" and "setting project knowledge goals" can lead to project knowledge management. In this phase, by transforming the knowledge goals into "Measurable Organizational Values (MOV)", business cases can be prepared and "knowledge creation" would be started. To make this documentary, the available (general and specific) knowledge in the knowledge base of an organization can be used (Leseure & Brookes, 2004). The new knowledge is created by combining the existing knowledge, coding and organizing the knowledge base and finally it is recorded and stored with the desired meta-data. In terms of cooperative and collaborative processes, inter-project and intra-projects, sharing and transferring knowledge transfer mechanisms (Ajith Kumar & Ganesh, 2009) and processes (Hanisch, Lindner, Mueller, & Wald, 2009) are used.

Planning a project
Up to this stage, the benefits and costs of the project have been clearly documented, objectives and project scope have been defined, project teams have been recruited and a formal project management office has been launched. Detailed plans are drawn up for the mandated activities, resource allocation and the controlling method for the next phase is determined. New plans are created and the acquisition of knowledge from them can be encoded and evaluated (Mitchell & Boyle, 2010). To make this documentary, the available (general and specific) knowledge on the knowledge base organizations can be used (Leseure & Brookes, 2004). New knowledge is created by combining the existing fields of knowledge, coding and organizing in the knowledge base it is recorded and stored with the desired meta-data. Sharing and transferring knowledge mechanisms and processes for cooperative and collaborative processes of inter-projects and intra-projects can be applied in this area (Liyanage, Elhag, Ballal, & Li, 2009;Hanisch, Lindner, Mueller, & Wald, 2009).

Executing a project
This phase includes the execution of activities defined in former phases. For this reason, this phase is the longest phase of the project. In this phase, the actual implementation and delivery of items are offered to gain the approval of the project stakeholders. Knowledge acquisition takes place among the items defined in the processes and document and knowledge will be used, evaluated and evolved to run the new experiences and will result in the creation of new knowledge (Mitchell & Boyle, 2010). In this phase, the available (general and specific) knowledge on the knowledge base of organizations can be used (Leseure & Brookes, 2004). New knowledge is created by combining the existing areas of knowledge, coding and organizing in the knowledge base followed by recording and storing with the desired meta-data. For cooperative and collaborative processes, interprojects and intra-projects, sharing and transferring knowledge transfer mechanisms (Liyanage, Elhag, Ballal, & Li, 2009) and processes (Hanisch, Lindner, Mueller, & Wald, 2009;Schindler, 2002) are used.

Monitoring and controlling a project
In order to ensure the "fulfillment of the requirements", the "quality of knowledge that is acquired, stored, distributed and applied in former steps", "project manager", "activities" and "resources and costs required for each item delivered during the implementation phase", stakeholders control and monitor the proper execution. To perform this phase, the available (general and specific) knowledge on the knowledge base an organization can be used (Leseure & Brookes, 2004). New knowledge is created by combining the existing fields of knowledge, coding and organizing in the knowledge base and finally recording and storing with the desired meta-data. To share and transfer knowledge mechanisms and processes for cooperative and collaborative processes, inter-projects and intra-projects can be used (Liyanage, Elhag, Ballal, & Li, 2009;Hanisch, Lindner, Mueller, & Wald, 2009).

Closing a project
This phase includes "presenting the final product delivered to customers (beneficiaries)", "knowledge of project documents", "terminating supplier contracts", "releasing project resources and receiving the project stakeholders' acceptance". To perform this phase, the available (general and specific) knowledge on the knowledge base of organizations can be used (Leseure & Brookes, 2004). Knowledge is acquired by coding and organizing the knowledge base and it is recorded and stored with the desired meta-data. Inter-projects and intra-projects and sharing and transferring knowledge transfer mechanisms (Liyanage, Elhag, Ballal, & Li, 2009) and processes (Hanisch, Lindner, Mueller, & Wald, 2009;Schindler, 2002) are used for cooperative and collaborative processes. One of the most important processes in this phase is "After Action Review" according to the most important "best practices" in the field of knowledge management projects. "After Action Review" should be practiced in any of the following circumstances: success/failure of the project sales project knowledge creation, capturing, acquisition, encoding and saving. Ultimately, the acquired knowledge can be shared and reused through mechanisms and technological components.

Step two-ontology evaluation
As mentioned before, the proposed ontology has been examined by two questionnaires in two steps regarding "Domain" and "Quality". Based on statistical analyses (Binomial and Mean tests), the proposed ontology has been accepted with 95% confidence with regard to both "Domain" and "Quality". By Friedman test with a confidence level of 95%, all three layers of ontology have been equal and homogenous. Cronbach's alphaindexwas96%in the first questionnaire and94% in the second questionnaire, then compared with70%alpha, it can be indicated that the validity of the questionnaires is high. The resulting assumptions outlined in methodology section will be described below:


Hypothesis 1: Experts' opinion in the first questionnaire will follow the normal distribution.
 Hypothesis 4: Experts' opinion in the second questionnaire will follow the normal distribution. Klmvgrf-Smirnov test results indicate a mismatch between the distribution data and the normal distribution. However, in "Domain" evaluation and in "Quality" evaluation, 11% and 7%of components follow the normal distribution respectively. Therefore, nonparametric tests (Ratio Test) were used to measure ontology and for other11% and 7% components, the parametric tests (Mean Test) were used.


Hypothesis 2: Domain of ontology is confirmed by experts.  Hypothesis 5: Quality of ontology is confirmed by experts.
For the majority of the components, the first hypothesis is rejected; then to measure the acceptance / rejection of "Domain" and "Quality" of the ontology, a Ratio Test is used. If all components of the hypothesis are confirmed, the final hypothesis asserting "The whole ontology is approved" will gain approval. The hypothesis would be rejected if all the components were rejected. Otherwise, the final judgment about the hypothesis will be difficult. In this study, Likert scale was used for the questionnaire. Therefore, this must be converted to an ordinal scale and the proportion can be defined as follows: "Completely agree" and "Agree" options: Ok "No Comment", "Disagree" and "Completely Disagree" options: Not ok Then the ratio of three options to five options is 0.6. If the ratio is less than 0.6, the number of people confirming the ontology would increase. Hence the i th hypothesis is as follows: H 0 : P i >= 0.6 i th component in the ontology is not approved (with respect to "Domain" and "Quality" points of view).
H 1 : P i < 0.6 i th component in the ontology is approved (with respect to "Domain" and "Quality").
According to the results of this test, the significance level is less than 0.05. Thus H0 will be rejected and H1 will be confirmed with 95% confidence. In addition, parametric and mean tests are used for the 11% and 7% of components that follow the normal distribution. The hypothesis is as follows: H 0 : μ> = 3 "Domain" and "Quality" of ontology are not approved. H 1 : μ< 3 "Domain" and "Quality" of ontology are approved.
Based on the results of descriptive statistics, the average for each component is smaller than three (Table 5, 6). Furthermore, in all 11% and 7% components, the significance level is smaller than 5%, which indicatesH0 rejection. Moreover, due to negative upper and lower levels of confidence, intervals can be determined with a confidence of 95%. Consequently, that H0is rejected and the average of expert opinions is smaller than three. Thus all components of the ontology were accepted with a confidence of 95% and the final hypothesis that "The whole ontology is approved" has been confirmed with 95% confidence regarding "Domain" and "Quality".  As previously mentioned, to measure the uniformity of the experts' agreement with the proposed ontology, the Friedman test is used. Then, following hypothesis tests are considered.
H 0 : There is no significant difference between experts' agreement on the layers of proposed ontology regarding "Domain" and "Quality". H 1 : There is a significant difference between experts' agreement on the layers of proposed ontology regarding "Domain" and "Quality". According to Table 7, the significant levels (0.170), (0.150) are larger than the error rate (0.05); therefore in the 95% confidence level, H0 hypothesis is accepted. The priorities of components in the domain ontology for project knowledge management based on the average ranking and analysis of variance using Friedman Test are mentioned. The smaller the average rating is the stronger endorsement the importance of those components would have. Based on Friedman test results, the experts' agreement on the ontology layers and its quality in different layers, have no significant difference, but with priority given to the test, it can be said that the layer of "People" needs further investigations compared to other layers and the first level (overall classification ontology based on the PPT pattern) in comparison with other layers has a stronger endorsement.

Conclusion
Given the importance of knowledge management and project-oriented approach to increase agility in organizations, having a strategic vision to these two categories is vital. Therefore, this study has presented domain ontology for project knowledge management in three layers: "People", "Technology" and "Process" with 115 cells.
The layer "People" has been divided into two subgroups: "Culture" and "Leadership", in 12 cells. The layer "Technology" has been classified into two subgroups of "Technology Component" and "Application", which has 72 cells and the layer "Process" that has been divided into five groups of "Initiating a Project", "Planning a Project", "Executing a Project", "Monitoring and Controlling a Project" and "Closing a Project" with 31 cells.
The main theoretical contribution of this study is an ontological framework linking Project Management and Knowledge Management, including two main parts: an ontology design, describing the key concepts related to project knowledge management and their inter-relationships (Fig. 7), and the evaluation of the ontology concerning domain and quality; which incorporates diversified issues for conducting project knowledge management from a competitive perspective.
Based on statistical analyses (Binomial and Mean Tests), the proposed ontology has been accepted with95% confidence and by Friedman test, three layers of which have been equal and homogenous.
At present, this ontology is in a proposal phase and needs further investigations in these areas:  Ontology creation: Using other patterns in succession to PPT pattern to design domain ontology for project knowledge management.  Applied ontology: transforming this ontology to one selected language and evaluating its efficiency in execution.  Improvement layers of ontology: Based on Friedman Test, research can improve the layer "People" in the future.  Implementing project knowledge management: This ontology can be used for decision-making in implementing project knowledge management.