Systematic Literature Review of the Implementation of Knowledge Codification Process 1 2 3 Franklin Espitia , Jenny Sánchez , Ernesto Galvis 1 Student of Master of Computing Systems Engineering, Universidad Nacional de Colombia, Bogotá, Colombia 2 Full Professor, Universidad Nacional de Colombia, Bogotá, Colombia 3 Associate Professor, Universidad del Magdalena, Santa Marta, Colombia 1
[email protected] 2
[email protected] 3
[email protected] Abstract: The purpose of this paper is to present the result of a Systematic Literature Review SLR of models, methodologies, guidelines, analysis, strategies, practices and frameworks related to the implementation of Knowledge Codification Process. The study was conducted following the SLR approach, taking into account only scientific documents published between 2004 and 2013. The revision consisted of four phases: search process, inclusion and exclusion filter, quality assessment, and data extraction, each of which was intended to select the most suitable information to help answer the research questions; case studies were mainly considered for future applications in organizations. The results of each document were identified as well as its activities, roles, tools, variables and indicators. The findings did not explicitly show documents where the implementation of Knowledge Codification Process was specified. As a result, based on the revision of the literature, the different components were taken from alternative processes. The results show first general aspects, presenting the number of documents per industry, per outcome and per research method used, followed by the list of activities, roles, tools, variables and indicators organized by the number of documents and by the type of outcome found in the revision. The result of the SLR can be a starting point to create an implementation model of Knowledge Codification Process, as well as give the readers an overview of the state of Knowledge Codification Process and promote Knowledge Management community in future investigations. Based on the result of the systematic review, there are no previous studies related to the implementation of Knowledge Codification Process. Keywords: Knowledge Management, Codification of Knowledge Management, Implementation of Knowledge Codification Process, Outlook of Codification of Knowledge Management, Organizations. 1. Introduction Knowledge is considered as the main competitive asset in a company. This can be categorized in two types: Explicit Knowledge, which is expressed in numbers or words and is shared in a formal and systematic way in books, magazines, manuals, publications, etc (McKenna, 2006), and Tacit Knowledge, which, according to Nonaka and Takeuchi, includes ideas, forms of thought, intuitions, among others (Nonaka and Takeuchi, 1995) (Aurum, Parkin and Cox, 2004). This last type of knowledge is more difficult to express and formalize, therefore, more difficult to share, and sometimes it can be communicated through the exchanging between individuals (Nonaka, Toyama and Konno, 2000)(Richardson et al., 2009). Therefore, Tacit Knowledge needs to be explicitly converted; in this process certain relevant knowledge could be lost (Richardson et al., 2009). In the literature reviewed, there was no formal definition for implementing a model of codification, nor a specification of what its components are. The definitions given by different authors of what the codification of knowledge is are presented as follows. For Awad and Ghaziri, Knowledge Codification, KC, is the conversion of tacit to explicit knowledge, with the purpose of being accessed, understood and reused by other people in the future (Awad and Ghaziri, 2004). A strategy of codification is proposed, where it should focus on capturing and codifying knowledge explicitly (documents, databases) and make it available for any person within the organization, for possible future reuse. Thus, the company invests only once in the development of explicit knowledge (storage), having it available for consultation as many times as required. As a result, this reuse avoids associated costs with ‘reinvention’ (Scheepers, Venkitachalam and Gibbs, 2004). Similarly, Samoilenko and Nahar regard KC as a phase of Knowledge Storage and consists of the conversion from tacit knowledge (personal experiences, skills and capabilities) into explicit knowledge (documents, tables, figures, databases, patents, manuals, etc) with the purpose of being applied by other members of the team (Samoilenko and Nahar, 2013). A way to codify tacit knowledge can be in documents created during the execution of procedures that were retained or also through meetings that ease the exchange of knowledge (Wood and Reynolds, 2013).
For Kraaijenbrink, Faran and Hauptman, KC process is the articulation and the transit of explicit knowledge from a human source to any other type of source. Once the knowledge is codified, this is detached from the initial source making it transferable independently for the community (Kraaijenbrink, Faran and Hauptman, 2006). In addition, Brown, Dennis and Gant indicate that, in a codification approach, organisations strongly depend on computers, carefully codifying the knowledge and storing it in documents of knowledge management systems with the purpose of making it accessible to a big number of people in the organisation. This approach is useful for organisations interested in standardizing the knowledge, and is centred around the exchange of knowledge through documents (Brown, Dennis and Gant, 2006). Other authors, like Ye, Marinova and Singh, note that KC refers to a unit process through which the articulated knowledge is converted into concrete knowledge, such as executable plans, work procedures and operative systems (Ye, Marinova and Singh, 2008). Unlike Arif, Egbu and Toma, for whom the Codification is the second step of Knowledge Retention, where tacit knowledge is converted into explicit knowledge (Arif, Egbu and Toma, 2010). For Rajalakshmi and Banu, KC means converting tacit knowledge in explicit knowledge, in a usable way for organisation members. This knowledge is sorted, categorised, indexed and stored in a repository so as to be shared and captured by a community. A portal or a blog can be used as codification tools of tacit knowledge (Rajalakshmi and Banu, 2012). For Hansen, Nohria and Tierney, the organisations usually expect to receive benefits from reusing the knowledge; with this approach, the role of IT is to support the storage and recovery of this kind of knowledge by people inside the organisation, whenever this is required (Hansen, Nohria and Tierney, 2000). In the above definitions, the authors emphasize both tacit and explicit knowledge. Moreover, some of them have some characteristics in common, such as storage and future use. Information Technology, IT, is taken into account by four authors, three of which mention the use of repositories. The definition proposed by Rajalakshmi y Banu covers the previous descriptions including sharing and converting. Therefore, this definition is taken as a foundation for this research as it is the most complete and more likely to cover the answers to the research questions. Along these lines, the purpose of this paper is to present the result of a Systematic Literature Review of scientific publications related to the implementation of Knowledge Codification Process. To achieve this, general aspects of all documents were analyzed, as well as the content of the outcome of each document. The analysis of general aspects was focused on the identification of the industry, the outcome, the type of publication and research method. The content analysis was intended to answer the following research questions: § What are the activities that comprise an Implementation Model of Knowledge Codification Process? § What are the roles that comprise an Implementation Model of Knowledge Codification Process? § What tools are needed to implement a model of Knowledge Codification Process? § What are the indicators and variables to measure the accomplishment of an implementation model of Knowledge Codification Process? To present the results of the SLR, the next structure was defined: In section 2, the methodology used for this SLR is described. The results answering research questions are described in section 3. Finally, the discussion and conclusions of the revision are presented in section 4. 2. Methodology The SLR is based on the method of Evidence-‐based research (Kitchenham et al., 2009); it consists in following a rigorous selection of published scientific papers through a search protocol with a defined scope and order of execution . The method consists of a search process in section 2.1, inclusion and exclusion criteria in section 2.2, quality assessment in section 2.3, and data extraction in section 2.4. The results are presented in section 3. 2.1 Search process Starting from the research questions, which were described in the introduction of this document, a first version of the search equation was defined, which was iteratively refined with different terms found in the initial revision, keywords and some synonyms, the final version of the search equation is described below:
TITLE-‐ABS-‐KEY(( ( knowledge W/0 ( accumulation OR adaptation OR adoption OR assembly OR assimilation OR capture OR codification OR collection OR combination OR compilation OR construction OR conversion OR creation OR documentation OR exteriorization OR integration OR organization OR presentation OR preservation OR refinement OR retention OR storage OR transformation) ) W/1 ( action OR activity OR aim OR approach OR assessing OR assessment OR capability OR diagnostic OR effectiveness OR evaluation OR framework OR goal OR guideline OR impact OR indicator OR measure OR method OR methodology OR metric OR model OR objective OR operation OR outcome OR output OR plan OR practice OR principles OR procedure OR process OR product OR program OR project OR proposal OR purpose OR result OR roadmap OR role OR scheme OR standard OR strategy OR system OR task OR technique OR tool OR variable )) OR ( ( codified OR explicit ) PRE/0 knowledge)) The search was limited to a range of 10 years, from 2004 to 2013; the equation was executed on the scientific database SCOPUS, where it is possible to consult different articles, papers, conferences, publications or book chapters. The total number of documents found with the search equation was 6946. 2.2 Inclusion and exclusion criteria The results were filtered by year, selecting only those which contained models, methodologies, guidelines, analysis, strategies, practices or frameworks. For this revision, the field of implementation was not taken into account with the purpose of having a greater range of evaluation. 1138 documents were obtained in this selection. A second filter was applied, selecting those documents which contained empiric works, such as interviews, surveys or case studies, related to Knowledge Management. From this filter 491 articles were obtained. 2.3 Quality assessment The articles were revised manually, excluding those which did not fulfil the quality criteria, those which did not have a well-‐defined methodology, or those which did not specify the model content or work method. At the end, a total of 53 papers were selected for data extraction. 2.4 Data extraction Each of the 53 documents selected was inserted in a relational database in order to have the option to query, generate extracts and read the obtained results in a friendly way. From the selected articles, the following data were collected: title, year, author(s), research method, outcome, industry, type of paper, Knowledge Management process and its components. For those not specifying any of previous information, general values were created. Regarding the presentation of the component results (activities, roles and tools), they were first grouped by the total number of papers in which they were mentioned. Also, they were cross-‐referenced in a table by outcome, specifying the number of matching documents for each outcome, and therefore determining the most common outcome with its most used components. With reference to the presentation of indicators and variables, they were classified by typology based on the definition offered by DANE (in Spanish: Departamento Administrativo Nacional de Estadística). They were associated to one activity found in the revision, (in case of having any ambiguity, the activity with more relation to the indicator or variable was chosen). Once the association was determined, the total number of indicators and variables by activity was added up to create a percentage distribution. With this result, the percentage of each activity was normalized in order to show in detail the total percentage of indicators and variables per activity. 3. Results In the SLR, no implementation models of Codification Knowledge Management Process were explicitly found, instead, different components were taken from the outcomes of alternate processes identified in the documentation. The results of the SLR are presented as follows: general aspects in section 3.1, the other results are presented in the same order as the research questions in the introduction of this paper. 3.1. General aspects From 53 papers selected, four general aspects were highlighted, the first being the industry. As it was mentioned in the initial search, a filter by industry was not done so as to have a wider range for the selection of potential eligible papers.
A total of 29 general papers did not specify the industry: 9 related to Software, 4 to Education, 3 to Construction, 2 to Economy, and Sport, Outsourcing, Judicial, Learning, Medicine and Mining with one paper. The second general aspect was the outcome that each document presented. The Model was the most common outcome with 24 articles, followed by Analysis with 11, Framework with 7, Methodology with 4, Guidelines with 3, Method with 2, Strategy and Practice with one related paper. The third general aspect was the type of publication; both articles in indexed journals and articles in conferences had the same number of documents with 25 results, and book chapters with 3 papers. The fourth and last general aspect was the research method used for the construction of the outcome, like model, methodology, guideline, etc. Case study was the most used with 36 papers, Survey with 14 documents, Interviews with 5, Analysis with 3 and Research with one related article. 3.2. List of activities The activities were selected based on the COMPETISOFT definition, where an activity is a group of specific tasks assigned to one or more roles for their completion (COMPETISOFT, 2008). Initially, 260 activities were selected apart from the process where they were mentioned. After debugging the registers and deleting the duplicates, 39 activities were obtained. Table 1 lists the total number of documents where they were identified, and those ones with 3 or more results. Retain and Transform only had one document, Interview, Integrate, Distribute, Protect, Review the repository and Explore had 2 documents. In Table 1, the activity Exchange with 24 papers is in the first place followed by Socialize with 22 and Store with 19 papers. Table 1: List of activities by the number of papers identified Activity Total of papers identified Exchange 24 Socialize 22 Store 19 Identify 15 Reuse 14 Give formal training – Acquire 11 Motivate 10 Standardize 8 Codify – Classify – Create 7 Evaluate – Document – Learn – Update – Internalize – Combine 6 Discuss – Apply 5 Capture – Experience feedback – Brainstorm – Create relationship and 4 trust – Externalize – Collect data – Index Collaborate – Transfer – Record 3 After cross-‐referencing the activities with the outcomes, the results showed that the most common outcome was Model, where its most used activities were Exchange with 13 papers, Store with 8 papers and Give Formal Training and Socialize with 7 papers. The second most common outcome was Analysis, where its most used activities were Socialize with 10 papers, Store with 7 articles and Exchange with 6 documents. The third most common outcome was Framework, where its most used activities were Identify with 6 articles, Motivate with 4 papers and Evaluate with 3 papers. The fourth most common outcome was Guideline, its most used activities being Acquire with 6 articles, Identify with 5 papers, Classify and Standardize with 2 documents. Other activities were associated to the outcomes but with a negligible number of papers. Strategy, Practice and Method are among the outcomes with fewer activities. 3.3. List of roles A Role is responsible for a group of activities of one or more processes. A role can be assigned to one or more people full or part time (Society, Bourque and Fairley, 2014) (COMPETISOFT, 2008) so that a person can take many roles. The selection process included those roles, positions or responsible the first place emphasized in the output of each paper. 107 registers were identified, after debugging and eliminating duplicated values, the final number of roles was established at 30. In Table 2, the results are listed accompanied by the number of papers in which they were identified. Table 2 lists only those roles with a result of 2 or more. First place is
occupied by Managers with 23 (among them there are high and medium ranks), human resources and business managers, followed by Areas or Departments and External Experts with 7 documents and Employees with 6. Reader, Sponsor, Storyteller, Labs users, Workgroups, Sales staff, Project team, Co-‐workers, Auditory and Research Department appeared each in one result. After cross-‐referencing the activities with outcomes, it was obtained that the most common outcome was the Analysis, where its most used roles were Managers with 13 papers, Departments (areas) with 6 documents, Junior Employess and Administrators with 3 articles. The second most common outcome was Model, where its most used roles were Managers with 4 papers and General Employees with 3 papers. The third most common outcome was Framework, where its most used roles were Managers and Content Team with 3 papers, followed by IT Positions with 2 documents. The fourth most common outcome was Methodology, where its most used roles were External Expert and Managers with 2 papers. Other roles were associated to the outcomes but with an unimportant number of papers. The outcomes Strategy and Guidelines were the ones with less roles associated, followed by Method and Practice that had no roles associated. Table 2: List of roles and number of papers identified Role Total of papers identified Managers 23 Departments (areas) – External Expert 7 General Employees 6 IT Positions – Administrator 5 Knowledge Management team – Content team – Junior Employee – 4 Contestant Community – Teachers – Students – Senior Employees 3 Research institutes – Trainers – Clients – Developers – Knowledge 2 enablers – Human Resources 3.4. List of tools A tool is one that aims to help implementing a task or activity easily and efficiently (Hlupic, Pouloudi and Rzevski, 2002). In the selection of tools, those based and not based on IT listed by the authors in the documents were taken. Initially 182 possible tools were registered, after debugging and removing duplicated registers, a total of 42 were obtained, which were distributed in two groups. The first group was composed of the tools not based on IT. 6 tools were identified altogether, where Meetings and Workgroups had the most associated papers with 5, Practice Communities with 4, Workshops and Formal Methodologies both with 2 and Seminary with one document. After cross-‐referencing the tools not based on IT with outcomes, the most common outcome was Analysis, where its most used tools were Workgroups with 3 papers, Practice Community with two and Seminary with one. The second most common outcome was Model, where its most used tools were Workgroups and Meetings both with one paper, and Formal Methodology with one article. The third most common outcome was Method, whose most used tools were Practice Community, Workshop and Meetings, all three with one result. Practice, Methodology and Guidelines were among the outcomes with fewer tools not based on IT. The outcome Strategy had no papers associated. The second group were tools based on IT. A total number of 36 were identified (Table 3), where Documents Repository with 14 and Databases with 10 had the largest number of papers associated. Table 3 lists only the tools with of 2 or more results. Checklists, Expert Systems, E-‐Services, Knowledge Portals, Draws, Genetic Algorithms, E-‐Portals and Web Servers returned one result only. After cross-‐referencing the tools based on IT with outcomes, the most common outcome was the Model, where its most used tools were Documents Repository with 5 papers, Databases, E-‐Learning, Emails, Intranet and Microsoft Office package with 4 papers. The second most common outcome was Analysis, where its most used tools were Documents Repository with 5 documents. Networking, Web Portal and Wiki with 2 results. The third most common outcome was Framework, where its most used tools were Databases with 3 articles. The fourth most common outcome was Methodology, where its most used tools were Standards and Printed
material, both with 2 papers. Other tools were associated to the outcomes but with a negligible number of papers. The outcomes Guideline, Method, Practice and Strategy had the lowest number of documents associated with one, two or three documents. Table 3: List of tools based on IT and number of papers identified Tool Total of papers identified Documents Repository 14 Databases 10 Audio-‐Video Technologies 7 Intranet 6 Microsoft Office – Email – Digital Library 5 Standards – Web Content – Data Mining – Wiki -‐ E-‐Learning 4 Online tools – Web Portal – Printed materials – Knowledge base – 3 Internet Networking – News – IT – Virtual Reality – E-‐meetings – Data 2 warehouses – Social networking (blogs, twitter) – Virtual communities – System support groups 3.5. Indicators and Variables An Indicator is a qualitative or quantitative expression observable, which allows to describe characteristics, behaviours or phenomena of a reality through the evolution of a variable or the relation between variables (DANE, 2005). A variable is one that represents all that can change in time (DANE, 2005). Initially 153 registers were identified, made of indicators and variables. After deleting duplicated registers and debugging information, 144 registers were obtained divided as follows: 55 indicators, 84 variables and 5 invalid registers. Different methodologies of how indicators and variables were applied were taken into account in order to evaluate the accomplishment of the outcome of each paper, yielding the following data: Questionnaire with 3 results, Interviews with 4 and Surveys with 3 papers. According to the distribution of the indicators by typology as indicated in the Guide for designing, building and interpretation of indicators (DANE, 2005), two types of typology were found: Measure with a total of 29 indicators, divided in Qualitative-‐Binary with 23, Qualitative-‐Categorical with 2 and Quantitative with 4 indicators. The second type was Intervention with a total of 26 indicators, classified in: Impact with 6, of Process with 10, Product with 7 and Outcome with 3 indicators. Team Effectiveness, Ease of accessibility to information, Frequency of meetings, Number of employees by unit, Technology Level, Standardized documentation, Flexibility of Knowledge Management were within the main indicators found in the SRL, among others. Efficacy, Autonomy, Culture, Information Volume, Risks, Team Size, Interaction between users, Leadership, Motivation, Grade of interest, etc. were within the main identified variables in the revision. Table 4 shows the percentage distribution of indicators and variables associated to activities. In the first place there is the activity Evaluate with a 21.6% of total indicators and variables, the activities Collaborate and Motivate with a 6.5% each, the activity Acquire, Innovate, Collect data and Experience Feedback with the experience with only 0.7% each. The results after normalizing the percentage distribution of indicators and variables by activity are presented in Figure 1. It shows how the activity Evaluate is distributed; of its total 21.6%, 70% are variables and 30% are indicators, Collaborate (78%, 22%), Motivate (44%, 56%). However, there are activities either with 100% of variables such as Innovate, Capture, Classify, or activities with 100% of indicators such as Update, Storage, Acquire, Collect Data, etc.
Table 4: Percentage distribution of activities by total of indicators and variables Activity Individual percentage of Total Percentage distribution of indicators and variables (%) indicators and variables (%) Evaluate 21.6 21.6 Collaborate – Motivate 6.5 13 Codify 5.7 5.7 Create 5 5 Update – Apply – Learn – Capture – 4.3 25.8 Classify – Discuss Store – Give formal training – 3.6 10.8 Document Recover 2.9 2.9 Combine – Identify – Reuse – Review 2.2 11 the repository – Socialize Standardize 1.4 1.4 Acquire – Innovate – Collect data – 0.7 2.8 Experience feedback
Figure 1: Percentage distribution of indicators and variables by Activity 4. Discussion and Conclusions As clear definition was not established for the components of an implementation model of Knowledge Codification Process. Based on the selected definition of Knowledge Codification provided by Rajalakshmi and Banu and the results of the SLR, a proposal for the components for an implementation model was put forward.
The components are presented in the same order as the research questions in the introduction of this paper. The activities obtained in section 3.2 were sorted by their frequency of occurrence and the definition mentioned above. A possible candidate list to compose an implementation model of Knowledge Codification Process is: Store, Reuse, Standardize, Codify, Classify, Evaluate, Document, Update, Index and Record. These cannot be seen 100% as activities, since they can be divided in a subcategory as tasks, which is a smaller unit of measurement. With regard to the roles that can comprise an implementation model of Knowledge Codification Process, based on the results, Manager is one of the most important with more papers identified. It shows how important this role is in different outcomes, either Model, Analysis, Framework, Methodology, etc. Other suggested roles are External Expert, Community, Management team, Trainers, Developers and Knowledge enablers. Concerning tools, if maintaining the same classification, the tools not based on IT are: Workgroups and Practice Community. For IT-‐based tools, these are: Documents repository, Wiki, Online tools, Virtual communities and System support groups. Regarding indicators and variables, a considerable number was found. They were classified by activity in order to give a clearer status of the existence of previous work of each one. It is possible to see how the activity Evaluate is distributed, from its total initial 21.6%, 70% are variables and 30% are indicators, showing the existence of previous work. Having more variables, it shows the desire to measure something but there are not enough indicators to show how, giving the possibility of delving and contributing more work with the proposition of new indicators. The same case can be argued for the activities Collaborate, Codify, Create, Apply, Learn, Give formal training, Reuse. Opposite case for activities like Update, Store, Review the repository, Acquire and Collect data, where a gap that exists in the presence of variables can be noticed; this means, the measurement of this activities is clear (what and how), thus, the contribution would not be substantial. The absence of implementation models in the results evidences that companies do not have a clear and easy structure to implement this type of organizational culture, underlining the need to create implementation models, not only for the Knowledge Codification process, but also for other processes, such as Identification, Transference, Protection, etc. Bibliography Arif, M., Egbu, C. and Toma, T. (2010) ‘Knowledge retention in construction in the UAE’, in, pp. 887–896. Available at: http://www.scopus.com/inward/record.url?eid=2-‐s2.0-‐ 84861052528&partnerID=40&md5=66cc0518b7db3d2e2bef411bdb300200. Aurum, A., Parkin, P. and Cox, K. (2004) ‘Knowledge management in software engineering education’, in. IEEE, pp. 370–374. doi: 10.1109/ICALT.2004.1357439. Awad, E. and Ghaziri, H. (2004) ‘Knowledge codification’, in Knowledge Management Systems. Prentice Hall. Brown, S. A., Dennis, A. R. and Gant, D. B. (2006) ‘Understanding the factors influencing the value of person-‐to-‐ person knowledge sharing’, in. doi: 10.1109/HICSS.2006.516. COMPETISOFT (2008) COMPETISOFT. Mejora De Procesos De Software Para Pequeñas Empresas. Available at: http://alarcos.esi.uclm.es/competisoft/ (Accessed: 18 May 2015). DANE (2005) Guía para Diseño, Construcción e Interpretación de Indicadores, Colombia Digital. Available at: http://www.dane.gov.co/files/planificacion/fortalecimiento/cuadernillo/Guia_construccion_interpretacion_in dicadores.pdf (Accessed: 9 October 2015). Hansen, M., Nohria, N. and Tierney, T. (2000) ‘What is your strategy for managing knowledge’, The knowledge management yearbook, 2001, pp. 55–69. Hlupic, V., Pouloudi, A. and Rzevski, G. (2002) ‘Towards an integrated approach to knowledge management:“hard”,“soft”and “abstract”issues’, Knowledge and Process Management, 9(2), pp. 90–102. Kitchenham, B. et al. (2009) ‘Systematic literature reviews in software engineering – A systematic literature review’, Information and Software Technology, 51(1), pp. 7–15. doi: 10.1016/j.infsof.2008.09.009. Kraaijenbrink, J., Faran, D. and Hauptman, A. (2006) ‘Knowledge Integration by SMEs – Framework’, in. Physica-‐Verlag HD, pp. 28–39.
McKenna, E. F. (2006) Business psychology and organisational behaviour: a student’s handbook. Psychology Press (4th Edition). Nonaka, I. and Takeuchi, H. (1995) ‘Global Organizational Knowledge Creation.’, in, pp. 197–223. Available at: http://search.ebscohost.com/login.aspx?direct=true&db=bth&AN=16017329&site=ehost-‐live. Nonaka, I., Toyama, R. and Konno, N. (2000) ‘SECI, Ba and Leadership: a Unified Model of Dynamic Knowledge Creation’, Long range planning, 33(1), pp. 5–34. Rajalakshmi, S. and Banu, R. S. D. W. (2012) ‘Analysis of tacit knowledge sharing and codification in higher education’, in 2012 International Conference on Computer Communication and Informatics, ICCCI 2012. doi: 10.1109/ICCCI.2012.6158906. Richardson, I. et al. (2009) ‘Knowledge Management in the Global Software Engineering Environment’, in. IEEE, pp. 367–369. doi: 10.1109/ICGSE.2009.57. Samoilenko, N. and Nahar, N. (2013) ‘IT tools for knowledge storage and retrieval in globally distributed complex software and systems development of high-‐tech organizations’, in 2013 Proceedings of PICMET 2013: Technology Management in the IT-‐Driven Services, pp. 1353–1369. Available at: http://www.scopus.com/inward/record.url?eid=2-‐s2.0-‐ 84890819796&partnerID=40&md5=c88303ec8ce49e57c8a9976a87cfb0a7. Scheepers, R., Venkitachalam, K. and Gibbs, M. R. (2004) ‘Knowledge strategy in organizations: Refining the model of Hansen, Nohria and Tierney’, Journal of Strategic Information Systems, 13(3), pp. 201–222. doi: 10.1016/j.jsis.2004.08.003. Society, I. C., Bourque, P. and Fairley, R. E. (2014) Guide to the Software Engineering Body of Knowledge (SWEBOK(R)): Version 3.0. 3rd edn. Los Alamitos, CA, USA: IEEE Computer Society Press. Wood, S. and Reynolds, J. (2013) ‘Knowledge management, organisational learning and memory in UK retail network planning’, Service Industries Journal, 33(2), pp. 150–170. doi: 10.1080/02642069.2011.614340. Ye, J., Marinova, D. and Singh, J. (2008) ‘Deliberate learning in the frontlines of service organizations’, in. Available at: http://www.scopus.com/inward/record.url?eid=2-‐s2.0-‐ 84858392001&partnerID=40&md5=9e0f5af295355c6804b852a428f25881.
