16 (2007) 806–815www.elsevier.com/locate/autcon

Automation in Construction

Developing a knowledge map for construction schedulingusing a novel approach

Jyh-Bin Yang ⁎

Institute of Construction Management, Chung Hua University, No. 707, Sec. 2, Wu-Fu Rd. Hsinchu 300, Taiwan

Accepted 8 February 2007

Abstract

A knowledge map is a vital tool for better knowledge management and learning. While application of knowledge maps in the constructiondomain remains in the initial stages of development, the construction industry is experience-oriented and therefore suited to knowledge maps. Thisstudy presents a novel approach for developing a knowledge map for construction scheduling. According to framework-based classification, thisstudy utilizes a science-specific search engine to search for literature on construction scheduling knowledge. Search results are then used todevelop a file cabinet knowledge map consisting of a contour map, and several trend and density charts. This map representation compensates forthe lack of various meanings in a single knowledge map. For novices interested in learning construction scheduling knowledge, results of thisstudy provide constructive information to know the key issues and research trends in the construction domain. In summary, this study presents asuitable procedure for extracting knowledge from public knowledge sources for development of a knowledge map. The proposed approach can beused for rapid generation of knowledge maps.© 2007 Elsevier B.V. All rights reserved.

Keywords: Knowledge map; Knowledge management; Construction management; Scheduling technique

1. Introduction

Schedule planning and control is a major task in successfulconstruction project management. Since the 1950s, the criticalpath method (CPM) and program evaluation and reviewtechnique (PERT) have been extensively adopted for projectscheduling and control. Excluding modifications on CPM andPERT, new robust scheduling approaches have not beendeveloped by academics and practitioners in the recent decades.Investigations to enhance the performance of available sched-uling techniques as necessary as requirements for researcherschange in the construction industry. Researchers or new learnersrequire an aid, a clear image of a study or learning target, as abasis for further study. The innovative concept of knowledgemanagement is a good choice.

Owing to the rapid evolution of the knowledge industry,knowledge engineers are confronted with the challenge of howto construct a well-linked knowledge network that allows a

⁎ Tel.: +886 3 5186684; fax: +886 3 5370517.E-mail address: jyhbin@chu.edu.tw.

0926-5805/$ - see front matter © 2007 Elsevier B.V. All rights reserved.doi:10.1016/j.autcon.2007.02.005

learner to acquire knowledge quickly. Many techniques havebeen developed to help the construction of knowledge net-works. A knowledge map that creates relationships amongisolated knowledge and represents knowledge via a hierarchystructure is a knowledge representation type that is mostvaluable. A knowledge map can clarify vague knowledge,enabling users and learners to easily find desired knowledge.Recently, domain knowledge about construction managementhas received considerable attention. Schedule management is animportant topic within construction management. When acomprehensive knowledge map for construction schedulingdomain can be constructed, domain development matures andlearners improve their knowledge.

Although approaches have been used to develop knowledgemaps, their usability for public knowledge sources is notproven. Knowledge engineers are required to test this usabilitybecause public knowledge sources (public search engines) willbe the most prosperous future knowledge sources. Furthermore,the knowledge map in the domain of construction scheduling,such as a hierarchy for project scheduling and monitoringproposed by Ahuja and Thiruvengadam [1], is insufficient for

807J.-B. Yang / Automation in Construction 16 (2007) 806–815

knowledge learners. One important functionality of a knowl-edge map is to enhance the quality of learning materials forpotential learners. This study tried to compensate for the lack ofvariety in a single type of knowledge map for constructionscheduling techniques.

The rest of this paper is organized as follows. Section 2 presentsa review of literature in the knowledge management andconstruction management domains to collect and summarize infor-mation on knowledge maps and construction scheduling tech-niques. Section 3 describes the data sources used for developing aknowledge map for construction scheduling techniques. Section 4presents the novel approach for rapidly constructing a knowledgemap. A knowledge map for construction scheduling techniques,including a knowledge framework with a contour map, and severaltrend and density charts, is organized on web pages with pictorialrepresentations. Section 5 presents conclusions and future researchdirections.

2. Literature review

2.1. Construction scheduling

Scheduling, as loosely defined by Sule, involves definingpriorities or arranging activities to meet certain requirements,constraints or objects [2]. Scheduling is a vital tool forsuccessful project management. For a construction project,project planning, particularly schedule planning, is at the heartof good project management because it provides the centralcommunication coordinating the work of all parties [3]. How tooptimally schedule a project is a basic skill of constructionmanagement professionals.

Betts and Lansley [4] reviewed all of the articles published inthe journal of ConstructionManagement and Economics (CME)from 1983 to 1992. They indicated that these articles publishedby CME are mainly concerned with production-related issues inthe construction industry. The topic of “project planning,scheduling and systems” is hot all through the analyzed time.Pietroforte and Stefani [5] reviewed all of the articles publishedin the ASCE Journal of Construction Engineering andManagement (CEM) between 1983 and 2000. They determinedthat the issue of time scheduling (including the subjects andtopics of cost/time scheduling, critical path method, line ofbalance technique, linear and vertical scheduling, networkplanning and analysis, other deterministic time schedulingtechniques, other nondeterministic time scheduling techniques,PERT and GERT, and time duration estimate, variability) werecommon during the investigation period. Simultaneously,Abudayyeh et al. [6] analyzed the research trends in theconstruction domain in the CEM journal from 1985 to 2002.They identified “scheduling” as the leading research topic. Thistopic has received considerable attention internationally with4.65% of the 879 articles analyzing scheduling-relatedproblems.

The domain of construction scheduling is an importantresearch topic. Furthermore, courses in construction planningand scheduling focus on construction scheduling techniquesand are core courses in construction management graduate

programs at universities worldwide. Learners must be given aclear picture of learning courses in the beginning of theirstudies. That is, comprehensive knowledge is a requisite duringlearning and studying. Application of knowledge managementusing a knowledge map meets this requirement.

2.2. Knowledge management

The Longman Dictionary of Contemporary English definesknowledge as the facts, information, skills, and understandingthat one has gained, especially through learning or experience[7]. Alavi and Leidner [8] considered knowledge from thefollowing five perspectives: state of mind; an object; a process; acondition of having access to information; and, a capability.Managing knowledge depends on the viewpoints of knowledgeengineers and end-users. Davenport and Prusak definedknowledge as a fluid mix of framed experience, values,contextual information and expert insight [9]. Knowledge existswhen different transformative processes, such as comparison,consequence, connection and conversation works. They alsoasserted that knowledge management is much more thantechnology; however, “techknowledge” is clearly a part ofknowledge management. Technology can diversify knowledgemanagement. With the rapid renovation of InformationTechnology (IT), knowledge of how to best manage knowledgeby IT is an important issue in academic and industrial domains.Recent studies examining knowledge management technolo-gies, applications and systems have been reviewed in detailelsewhere [8,10].

In the construction industry, engineering/management con-sulting firms are knowledge-based companies whose primaryproduct is knowledge. Knowledge management in such firms isidentical to that in general manufacturing or service enterprises,and has recently attracted intensive study. Other firms regardknowledge management as a competitive advantage rather thana survival tool. This circumstance results in an industry crisis.The major difficulty in implementing knowledge managementin the construction industry is formulation and implementationof a strategy [11]. Although previous studies attempted to selector to develop an appropriate strategy for the constructionindustry [11,12], managerial courage is required to face thischallenge and achieve changes.

Knowledge management associated with learning construc-tion scheduling techniques in this study is an issue of how to bestprovide access to required knowledge via a user-friendlyinterface. Several accepted methods exist for knowledge repre-sentation, including rules, frames, semantic networks, conceptdiagrams (concept mapping) and knowledge maps (knowledgemapping) [13,14]. The knowledge map method, a navigation aidto explicit and tacit knowledge [15], meets this study'srequirements.

2.3. Knowledge map

Davenport and Prusak defined a knowledge map as aknowledge yellow pages or a cleverly constructed database [9].A knowledge map, which can be used to point to knowledge,

Fig. 1. Knowledge map development approach.

808 J.-B. Yang / Automation in Construction 16 (2007) 806–815

does not contain knowledge. Knowledge maps are created bytransferring certain aspects of knowledge into a graphic form thatis easily understandable by end-users [16]. That is, knowledgemaps are consciously designed communication mediums be-tween map makers and map users [17]. A knowledge map playsimportant roles in implementing knowledge management. Allcaptured knowledge can be summarized and abstracted through aknowledge map [14]. Knowledge maps gather explicit andcollected knowledge which can be shared, and facilitateemergence of tacit knowledge of new relationships. Effectiveknowledge mapping brings map users several returns includingeconomic, structural, organizational culture and knowledgereturns [17]. Moreover, partial knowledge returns acceleratelearning curves by helping users locate an effective route, plan,and scenario or sequence of actions, and assist participants incommunicating with others regarding the new relationships andideas using a shared vocabulary. Partial knowledge returns alsohelp users identify new areas in the emerging search for actionableinformation. Therefore, the benefit of knowledge returns isachievable when a knowledge map for construction schedulingtechniques is constructed.

Several representational methods exist for knowledge maps,including the file cabinet knowledge map, cognitive map,knowledge networking chart, scatter chart, contour chart, listchart, category map [18], topic map [19] and concept map [20].This study uses the simple file cabinet knowledge map as thebasis for knowledge management in relation to constructionscheduling. The main cause of using the method is due to itsfriendly to use and easy to read for users.

The key issue in developing a knowledge map is locatingimportant knowledge and then organizing this information.Several approaches exist for knowledge map development.Rouse et al. [21] proposed a procedure for constructing aknowledge map for R&D/technology management. Theirprocedure consists of the following six steps: extraction ofknowledge; compilation of knowledge; derivation of assertions;sorting and labelling; representation of relationships; interpre-tation; and, iteration. Moreover, Kim et al. proposed a procedurefor building a knowledge map that is useful for developing aknowledge management system for an organization [14].

However, available approaches for knowledge map devel-opment are not targeted to public knowledge sources. Thisstudy proposes a novel procedure that is suitable for extractingknowledge from public knowledge sources when developing aknowledge map. The proposed procedure has the following sixsteps: defining organizational knowledge; process map analy-sis; knowledge extraction; knowledge profiling; knowledgelinking; and, knowledge map validation. Fig. 1 shows thedetailed procedure.

3. Data sources

3.1. Database introduction

The Internet has several powerful search engines, includingGoogle™, AllTheWeb, Yahoo®, TeomaSM, and AltaVista™[22]. Finding specific resources (e.g., refereed papers) for a

scholarly use is difficult using these general search engines. Forresearchers, specialty search engines such as Scirus, Google™Scholar and Internet Archive are relatively more convenient.Scirus is a comprehensive science-specific Internet searchengine that was voted the best specialty search engine in 2001and 2002 [22]. In addition to web pages, Scirus indexes thefollowing special sources: 14.6 million MEDLINE® citations,5.5 million ScienceDirect® full-text articles, 1.2 million patentsfrom the USPTO, 261,000 e-prints on ArXiv.org, 5,352BioMed Central full-text articles, 5,352 BioMed Central full-text articles, 10,600 NASA technical reports and 14,878 full-text articles from Project Euclid [23]. The ScienceDirect®Online database is a subscription information source forscientific, technical and medical research that offers access tomillions of articles from over 1800 journals, including popularconstruction-specific journals (such as Automation in Con-struction, Building and Environment and International Journalof Project Management), is an ideal database for exploration.This study uses the Scirus search engine to survey existingliterature regarding construction scheduling techniques.

3.2. Search criteria

Scirus search engine provides a comprehensive interface thatallows users to search web-based and journal-based sources.Although web pages may include significant amounts of non-peer-reviewed information, web-based data is excluded in thisstudy. This study uses the following search criteria.

• Published years: The database covers the period from 1920to now. Owing to the fact that information is scarce duringthe early years, the search period was divided unequally. Theyears 1920–1970 are organized as an independent timeframe, while other years are organized into individual 5-yeartime frames.

809J.-B. Yang / Automation in Construction 16 (2007) 806–815

• Information types: The database includes the types ofabstracts, articles, books, company homepages, conferences,patents, preprints, and scientist homepages. Althoughjournal article information may not include all informationtypes, this study uses all types and excludes non-journalarticles.

• File formats: The database contains two types of file formats,HTML and PDF. This study searches for journal articlesdisplayed in HTML or as PDF.

• Content sources: The database includes articles from thefollowing journal sources: BioMed Central, CrystallographyJournals Online, MEDLINE, Project Euclid, ScienceDirect,Scitation, Society for Industrial & App. Mathematics, E-Print ArXiv, CogPrints, NASA, US Patent Office, EuropeanPatent Office, Japanese Patent Office, and Patent Coopera-tion Treaty Office. This study uses all sources to maximizethe amount of knowledge contained in the search space.

• Subject areas: Domain areas in the database includeagricultural and biological sciences, astronomy, chemistryand chemical engineering, computer science, earth andplanetary sciences, economics, business and management,engineering, energy and technology, environmental sciences,languages and linguistics, law, life sciences, materialsscience, mathematics, medicine, neuroscience, pharmacolo-gy, physics, psychology, social and behavioral sciences, andsociology. This study adopts the areas of economics,business and management and engineering, energy andtechnology as target areas.

3.3. Knowledge exploration processes

The purpose of this knowledge exploration procedure is tofind records related to construction scheduling in the database.Although the employed search engine provides comprehensivesearch approaches, a simplified search process (focusing onkeyword searching) was used to retrieve desired publicationinformation. The knowledge exploration processes in the web-based database are summarized as follows.

1. Setting basic criteria described in the Section 3.2.2. Entering a keyword (the knowledge field) listed in Table 1.

Table 1Construction scheduling-related records

Knowledge field 1920–1970 1971–1975 1976

(1) Simulation 0 6 39(2) Delay analysis 1 3 8(3) Resource-constrained scheduling 0 1 1(4) Critical path method 0 0 1(5) Resource leveling 0 0 0(6) Program evaluation and review technique 0 0 0(7) Time-cost trade-off 0 0 0(8) Graphical evaluation and review technique 0 0 1(9) Network planning and analysis 0 0 0(10) Linear scheduling techniques 0 0 0(11) Critical chain scheduling 0 0 0Sum 1 10 50

3. Running a search scenario to extract knowledge in thesystem.

4. Checking search results to eliminate unsuitable recordsbased on author's judgement. Eliminated records includescheduling techniques not used in the construction industry,and those with matching keywords not desired forscheduling techniques.

5. Saving refined records in a database software system. Thesaved information includes title, author(s) and sources forany found article.

4. Approach for knowledge map development

Knowledgemanagement is an emerging field that has receivedconsiderable attention from research and industry communities.However, how to best exhibit the explored knowledge remainscontroversial. The most popular knowledge representationalapproach is the knowledge map, which can pictorially depictabsorbed knowledge to help users in digesting and learning theknowledge more quickly and conveniently than others. Based onthe approaches proposed by two studies [14,21], this studyemployed a six step approach, shown as Fig. 1, to develop aknowledge map for construction scheduling techniques. Theproposed approach is as follows.

4.1. Knowledge framework establishment

This study collected several textbooks [3,24–27] regardingconstruction scheduling and listed all tables of contents asoriginal sources for developing a knowledge frameworkconsisting of 23 construction scheduling techniques. To covernew or advanced issues of construction scheduling, this studyalso reviewed numerous articles [1,5,28–35] related toconstruction scheduling to induce a comprehensive framework.Fig. 2 shows the proposed framework of construction schedul-ing techniques.

4.2. Knowledge sources determination

This study used the Scirus search engine as the knowledgeexploration source. Fig. 3 shows the number of records found

–1980 1981–1985 1986–1990 1991–1995 1996–2000 2001–2004

51 78 114 158 12816 23 38 90 772 7 21 42 231 10 5 17 90 2 1 8 50 1 2 8 30 2 3 2 31 1 4 0 10 2 0 2 12 1 0 0 10 0 0 0 273 127 188 327 253

Fig. 2. Knowledge framework of construction scheduling techniques.

810 J.-B. Yang / Automation in Construction 16 (2007) 806–815

using “scheduling” and “construction and scheduling” askeywords. The average percentage of the number of recordsfound with “construction scheduling” compared with that foundwith “scheduling” is about 15%. The Scirus search enginecontains general domain publications, namely, there are manyjournal sources included in the system. Whereas, the value of15% implies there are certain records that can be found if theprevious setting is kept. Obviously, the knowledge of construc-tion scheduling in the Scirus search engine is extremelyattractive to researchers. The following process (knowledge

Fig. 3. Data profile of sc

extraction) is based on records obtained using “constructionscheduling” as keywords.

4.3. Knowledge extraction

For elaborating the information in the knowledge source, thekeywords (knowledge field) in Table 1 are used to mine crudeknowledge. Table 1 lists the detailed record numbers obtained,whereas Table 2 lists the sum value from Table 1, the number offound records using “construction scheduling” keywords, andthe value of the Significant Index for studied periods. Eq. (1)shows the calculation algorithm for the Significant Index that iscalculated using the sum of found records via the individualknowledge field (Rindi) divided by the records found via the“construction scheduling” keyword (Rcs).

Significant Index ¼ Rindi

Rcs� 100%: ð1Þ

For instance, between 1981 and 1985, 128 records are foundby Scirus using “construction scheduling” as keywords.Simultaneously, 51, 16 and 2 records are found using“simulation,” “delay analysis” and “resource-constrainedscheduling” as keywords based on 128 records. Table 2shows the Significant Index value for each period studied.Obviously, the sum of the number of records obtained usingindividual knowledge fields is not equal to the number obtainedusing “construction scheduling” (Table 1). If further analysis isbased on information in Table 1, the search results aresignificant because the average Significant Index value isN60%. Therefore, the following processes are analyzed basedon the records listed in Table 1.

4.4. Knowledge compilation

Following knowledge extraction, the proposed processcompiles mined knowledge in a pictorial form to enhance thetransparency of knowledge for users. For instance, Fig. 4 showsa contour map displaying all extracted knowledge in ameaningful map that allows users to decipher the variations inevery knowledge field at different time frames and to read thecorrelations between any two fields.

heduling techniques.

Table 2Significant Index values for studied periods

Feature 1920–1970 1971–1975 1976–1980 1981–1985 1986–1990 1991–1995 1996–2000 2001–2004

Sum of the number of individualknowledge field (Rindi)

1 10 50 73 127 188 327 253

Number of found constructionscheduling records (Rcs)

13 17 75 128 166 271 529 391

Significant Index 7.7% 58.8% 66.7% 57.0% 76.5% 69.4% 61.8% 64.7%

811J.-B. Yang / Automation in Construction 16 (2007) 806–815

For each knowledge field, this process was also used toassemble obtained information and show its characteristics. Forinstance, a trend chart (Fig. 5) exhibits the trends for a specificknowledge field (the “simulation” technique), whereas a densitychart (Fig. 6), exhibits the publication density of a specificknowledge field (the “resource-constrained scheduling” tech-nique) during different time frames. The trend chart alsoprovides users with an image of the trend in studied knowledge.For instance, the long-term trend of the “simulation” techniquebecame increasingly popular over time (Fig. 5).

Conversely, the density chart identifies the hottest periodregarding studied knowledge. For instance, Fig. 6 showsresearch density of the “resource-constrained scheduling”technique. Clearly, the time frame of 1996–2000 is the hottestperiod, during which 50 records are found in Scirus, a value thatis roughly 10% of all records found using the “resource-constrained scheduling” keyword phrase. In summary, thisknowledge compilation process renders extracted knowledgereadable; by using these charts, knowledge users can judge thevalue of interesting field.

4.5. Knowledge representation

This study made a pictorial representation of extracted andcomplied knowledge. All knowledge was organized using aweb page editing system. Therefore, knowledge users can easilyreach desired knowledge via any web browser. For eachknowledge field, trend and density charts were created to depict

Fig. 4. Contour map for construction scheduling techniques.

implicit knowledge. Additionally, a database for accessing therecords was created to depict explicit knowledge (obtainedjournal articles of construction scheduling techniques). Fig. 7shows a snapshot of organized knowledge viewed via a webbrowser.

4.6. Knowledge interpretation

This study created a knowledge map for constructionscheduling techniques. Through the proposed processes, amore mature knowledge map (Fig. 8) than traditional file cabinetknowledge map was constructed. This map helps map usersunderstand the development history of construction schedulingtechniques. The proposed or publication years of discussedtechniques in the construction scheduling domain are in the map.Furthermore, this map also provides a detailed classificationframework of construction scheduling techniques. Specifically,the map provides users with a comprehensive understanding ofeach scheduling technique. This study transferred the map into aformat of a file cabinet knowledge map (left in Fig. 7).Map userscan easily browse the desired knowledge and view the compliedknowledge chart. Moreover, based on the contour map (Fig. 4),users can identify popular scheduling techniques; for instance,the simulation technique in construction scheduling has recentlyreceived considerable attention.

A trend chart shows development tendencies within eachknowledge field. The trend for simulation technique is graduallyincreasing (Fig. 5). A density chart indicates a research focusduring different time frames. For example, the resource-constrained scheduling technique (Fig. 6) has the largest ratio(the resource-constrained scheduling to all construction sched-uling techniques found) during the period of 1996–2000,meaning that during this period, the ratio of obtained articlesfor resource-constrained scheduling to all construction

Fig. 5. Trend chart for simulation technique.

Fig. 6. Density chart for resource-constrained scheduling technique.

812 J.-B. Yang / Automation in Construction 16 (2007) 806–815

scheduling records is 9.45%, the largest among all time framesanalyzed. Furthermore, the figure also reveals that the den-sities of the resource-constrained scheduling technique varyamong time frames, and the time frame with the largestconcentration is 1996–2000. For this period, 50 research recordswere found.

Obviously, the trend chart shows the direction of research,and the density chart shows the relative ratio and degree ofconcentration of research for a particular technique.

5. Conclusion and recommendations

Knowledge of construction management (or ProfessionalConstruction Management) has received considerable attention

Fig. 7. Data access

in the construction industry. Scheduling knowledge is the mostessential issue owing to the fact that it is at the core ofconstruction management, especially for project planning andcontrol. Constructing a well-rounded knowledge map forconstruction scheduling techniques can make domain knowl-edge development more matured and provide learners withmore comprehensive knowledge. This study generated a mainclassification of construction scheduling knowledge usingseveral textbooks, and then employed the Scirus, a science-specific search engine, to search for the literature aboutconstruction scheduling techniques for generating a knowledgemap. Based on the knowledge map development approachemployed in this study, several valuable lessons were learned.

• Although the knowledge map constructed is not compre-hensive as this is a pilot study based on a single web-basedsearch engine, the map provides a good reference forlearning construction scheduling techniques.

• The knowledge map development approach employed in thisstudy was successfully examined for constructing a knowl-edge map based on a public search engine. This approachwill be valuable for similar studies that use searchabledatabases to construct knowledge maps.

• Knowledge sources determine the value of extracted knowl-edge. This study used a public search engine to construct aknowledge map. This search engine is not specially designedfor the construction industry. This circumstance results inthe outcome of this study having limited reliability. Incorpo-rating additional databases containing increased numbers of

to rough data.

Fig. 8. Knowledge map for construction scheduling techniques.

813J.-B. Yang / Automation in Construction 16 (2007) 806–815

construction-related journals will increase knowledge mapreliability.

• Complied knowledge lacks balance for previous studiesinvestigating different fields of construction schedulingknowledge. For instance, from 1996–2004, literature focusedon “simulation” and “delay analysis' knowledge fields. Other

fields received litter attention. A mature industry or researchissue requires diverse inputs to improve its outputs. That is, allscheduling fields require increased contributions fromresearchers, not just the hottest field.

• Knowledge map construction requires a knowledge engineerfamiliar with both domain knowledge and knowledge

814 J.-B. Yang / Automation in Construction 16 (2007) 806–815

regarding employed technology. Developing a knowledgemap in the construction industry requires knowledgeengineers with a construction industry background. Suchengineers are rare. However, if construction professionalscan improve their ability to manage knowledge, knowledgemanagement in the construction industry will improve.

• Knowledge management is a non-resisting trend for allindustries. The construction industry, frequently regarded asa traditional industry, has few stories regarding successfulknowledge management. Promoting construction expertiseis essential to acquire knowledge of knowledge managementin a scenario of rapid advances in industry innovation.

The most elaborate knowledge maps for construction schedul-ing learners can be quite complex due to changes in knowledgeover time. Research investigating scheduling, even that forconstruction scheduling, remains extremely popular. A compre-hensive knowledgemapmust be developed for subsequent studies.This pilot study attempts to develop a preliminary knowledgemapusing a web-based search engine. Some recommendations basedon this study are provided for further research.

• Since the knowledge map is based on a single search enginethat accesses a finite number of construction-related journals,convincing map users the map's integrity is difficult. Furtherresearch is needed that accesses additional construction-related journals to improve the comprehensiveness of theknowledge map.

• Although the knowledge map is quite simple because severalstructured and popular digital databases, such as the ISI Webof KnowledgeSM (including SCI, SSCI andA&HCI indexes),ProQuest's ABI/INFORM, ICONDA (The InternationalCONstruction DAtabase) and ASCE's CE Database, arenot included. This study is a good step toward development ofa perfect map. Inclusion of additional databases is necessaryto construct a comprehensive knowledge map.

• The returns a knowledge map offers map users must beevaluated. A feasible evaluation methodology must bedeveloped to justify construction-related maps.

A knowledge map is the knowledge yellow pages created bytransferring certain aspects of knowledge into a graphic form.This study proposed a novel approach for quickly generatingknowledge maps. The proposed approach was demonstrated bygenerating a knowledge map for construction scheduling tech-niques. A file cabinet representation was combined with severalmeaningful charts that compensate for the lack of variety ofmeanings in a single knowledge map type. For those interestedin construction scheduling knowledge, results of this studyprovide constructive information to identify the key issues andresearch trends in the area, and, thus, provide a basis for furtherresearch.

Acknowledgments

The author would like to thank the National Science Council,Taiwan, ROC, for partial financial support of this research under

Contract Nos. NSC92-2211-E-216-011-CC3 and NSC93-2211-E-216-012-CC3. The author would also like to thank Mr. Kun-HungWu, amaster of the Institute of ConstructionManagement,Chung Hua University, for helping with related information andexecuting necessary web searches for this research.

References

[1] V. Ahuja, V. Thiruvengadam, Project scheduling and monitoring: currentresearch status, Construction Innovation 4 (1) (2004) 19–31.

[2] D.R. Sule, Industrial Scheduling, PWS Publishing Company, Boston, MA,1997.

[3] G.D. Oberlender, Project Management for Engineering and Construction,McGraw-Hill, Boston, MA, 2000.

[4] M. Betts, P. Lansley, Construction management and economics: a reviewof the first ten years, Construction Management and Economics 11 (4)(1993) 221–245.

[5] R. Pietroforte, T.P. Stefani, ASCE Journal of Construction Engineering andManagement: review of years 1983–2000, Journal of ConstructionEngineering and Management, ASCE 130 (3) (2004) 440–448.

[6] O. Abudayyeh, A.D.-D. Young, E. Jaselskis, Analysis of trends inconstruction research: 1985–2002, Journal of Construction Engineeringand Management, ASCE 130 (3) (2004) 433–439.

[7] Longman Group, Longman Dictionary of Contemporary English, Long-man Group UK Limited, Harlow UK, 1991.

[8] M. Alavi, D.E. Leidner, Review: knowledge management and knowledgemanagement systems: conceptual foundations and research issues, MISQuarterly 25 (1) (2001) 107–136.

[9] T.H. Davenport, L. Prusak, Working Knowledge: How OrganizationsManage What They Know, Harvard Business School Press, Boston, Mass,1998.

[10] S.H. Liao, Knowledge management technologies and applications—literature review from 1995 to 2002, Expert Systems with Applications25 (2) (2003) 155–164.

[11] H.S. Robinson, P.M. Carrillo, C.J. Anumba, A.M. Al-Ghassani, Develop-ing a business case for knowledge management: the IMPaKT approach,Construction Management and Economics 22 (7) (2004) 733–743.

[12] J.M. Kamara, C.J. Anumba, P.M. Carrillo, A CLEVER approach toselecting a knowledge management strategy, International Journal ofProject Management 20 (3) (2002) 205–211.

[13] J.L. Gordon, Creating knowledge maps by exploiting dependent relation-ships, Knowledge-Based Systems 13 (2–3) (2000) 71–79.

[14] S. Kim, E. Suh, H. Hwang, Building the knowledge map: an industrial casestudy, Journal of Knowledge Management 7 (2) (2003) 34–45.

[15] D. Grey, Knowledge mapping: a practical overview, SWS Journal (1999)available at: http://smithweaversmith.com/knowledg2.htm, accessed 31January 2005.

[16] P.-H. Speel, N. Shadbolt, W. de Vries, P.H. van Dam, K. O'Hara,Knowledge mapping for industrial purposes, 12th Workshop onKnowledge Acquisition Modeling and Management, Alberta, Canada,1999, Available at: http://sern.ucalgary.ca/ksi/KAW/KAW99/papers/Speel1/index.html#2.Knowledge_Mapping, accessed 31 January 2005.

[17] M. Wexler, The who, what and why of knowledge mapping, Journal ofKnowledge 5 (3) (2001) 249–263.

[18] Construction and Planning Agency, A study on Developing ClassificationFramework for Construction Knowledge Map and Knowledge ExchangeStandard (Final Report, in Chinese), Construction and Planning Agency,Ministry of Interior, Taiwan (ROC), 2003.

[19] TopicMaps.Org, Topic Maps Paradigm. Accessed at: http://www.topic-maps.org/ (accessed 31 January 2005).

[20] R. McAleese, The knowledge arena as an extension to the concept map:reflection in action, Interactive Learning Environments 6 (3) (1998)251–272.

[21] B.W. Rouse, S.B. Thomas, R.K. Boff, Knowledge maps for knowledgemining: application to R&D/Technology Management, IEEE Transactionson Systems, Man and Cybernetics. Part C, Applications and Reviews 28 (3)(1998) 309–317.

815J.-B. Yang / Automation in Construction 16 (2007) 806–815

[22] Search Engine Watch, Search Engine Watch Awards, 2002 Available at:http://searchenginewatch.com/awards/article.php/2155921, 2003, accessed31 January 2005.

[23] Elsevier, Scirus search engine. Accessed at: http://www.scirus.com/,accessed 31 2005.

[24] E.M. Willis, Scheduling Construction Projects, John Wiley & Sons, NewYork, NY, 1986.

[25] M.T. Callahan, D.G. Quackenbush, J.E. Rowings, Construction ProjectScheduling, McGraw-Hill, Boston, MA, 1992.

[26] J.W. Hinze, Construction Planning and Scheduling, Prentice-Hall, NewJersey, 1998.

[27] R.J. Clough, G.A. Sears, S.K. Sears, Construction Project Management,4th ed. John Wiley & Sons, New York, NY, 2000.

[28] S. Alkass, M. Mazerolle, F.C. Harris, Construction delay analysistechniques, Construction Management and Economics 14 (5) (1996)375–394.

[29] J. Gordon, A. Tulip, Resource scheduling, International Journal of Project15 (6) (1997) 359–370.

[30] R.B. Harris, P.G. Ioannou, Scheduling projects with repeating activities,Journal of Construction Engineering and Management, ASCE 124 (4)(1998) 269–278.

[31] L.P. Leach, Critical chain project management improves projectperformance, Project Management Journal 30 (2) (1999) 39–51.

[32] W. Herrlelen, R. Leus, On the merits and pitfalls of critical chainscheduling, Journal of Operations Management 19 (5) (2001) 559–577.

[33] H. Steyn, Project management applications of the theory of constraintsbeyond critical chain scheduling, International Journal of ProjectManagement 20 (1) (2002) 75–80.

[34] K.D. Gothand, Schedule delay analysis: modified windows approach, CostEngineering 45 (9) (2003) 18–23.

[35] I.M. Mahdi, A new LSM approach for planning repetitive housingprojects, International Journal of Project Management 22 (4) (2004)339–346.