Enhancing knowledge exchange through web map-based knowledge management system in construction: Lessons learned in Taiwan

Automation in Construction 15 (2006) 693 – 705 www.elsevier.com/locate/autcon

Enhancing knowledge exchange through web map-based knowledge management system in construction: Lessons learned in Taiwan Yu-Cheng Lin a,*, Lung-Chuang Wang b, H. Ping Tserng c a Institute of Civil Engineering and Disaster Reduction Technology, Ching Yun University, No. 229, Chien-Hsin Rd., Jung-Li, Taiwan Institute of Civil and Disaster Prevention Engineering, National Taipei University of Technology, No. 1 Chung-Hsiao E. Rd., Sec. 3, Taipei, Taiwan Division of Construction Engineering and Management, Department of Civil Engineering, National Taiwan University, No. 1 Roosevelt Rd., Sec. 4, Taipei, Taiwan b

c

Abstract Knowledge management involves creating, securing, coordinating, combining, retrieving and distributing knowledge. Knowledge can be reused and shared among engineers and experts to enhance construction processes and decrease the time and cost of solving problems. This study presents a novel and practical method to capture and represent construction project knowledge by using network knowledge maps. Network Knowledge Maps (NKM) gives users an overview of available and missing knowledge in core project areas, enabling tacit and explicit knowledge to be managed appropriately. This study addresses application of knowledge management in the construction phase of construction projects, and presents a construction Map-based Knowledge Management (MBKM) concept and system for contractors. The MBKM system is then utilized in selected case studies involving a High-Tech factory building enterprise in Taiwan to verify the proposed methodology and indicate the effectiveness of sharing knowledge, particularly in the construction phase. Knowledge can be captured and managed to benefit future projects by effectively utilizing information and web technologies during the construction phase of a project. The results of this study demonstrate that an MBKM-like system can be applied effectively in knowledge management systems in the construction industry by using map-based knowledge management and web technology. D 2005 Elsevier B.V. All rights reserved. Keywords: Knowledge management; Knowledge map; Web-based application; Construction projects

1. Introduction Referring to the organization, creation, sharing and distribution of knowledge within organizations, knowledge management (KM) can take the form of idea management systems enabling employee ideas and suggestions to be captured and shared online. Many organizations are currently involved in KM efforts to apply knowledge effectively both within their organization and externally to their stakeholders and customers [1,2]. Knowledge is the true asset of a marketing-oriented organization, and its integration across departments and disciplines should be emphasized [3]. Reusing information and knowledge minimizes the need to refer explicitly to past projects, reduces the time and cost of

* Corresponding author. E-mail addresses: [email protected] (Y.-C. Lin), [email protected] (L.-C. Wnag), [email protected] (H.P. Tserng). 0926-5805/$ - see front matter D 2005 Elsevier B.V. All rights reserved. doi:10.1016/j.autcon.2005.09.006

solving problems, and improves the quality of solutions during the construction phase of a construction project. Sharing experience and knowledge means that problems in construction projects do not need to be repeatedly solved. Reducing problem-solving reduces costs and the probability of repeat problems. Several enabling activities should be considered to help attain the eventual goal of efficient experience and knowledge reuse; experience and knowledge should be preserved and managed, restated activities should be captured, modeled, stored, retrieved, adapted, evaluated and maintained [4]. Kamara [5] reviews various KM initiatives to assess the extent of KM implementation in the AEC industry, evaluates current KM strategies in AEC firms from two research projects, and indicates that effective knowledge management combines mechanistic and organic approaches incorporating both technological and organizational issues. Currently, enterprises in the AEC industry have successfully collected and stored explicit information in enterprise databases, but

694

Y.-C. Lin et al. / Automation in Construction 15 (2006) 693 – 705

have not always been successful at tacit knowledge retrieval and sharing [6]. A case investigation from the oil and gas industry is adopted to explore the KM activities of eight leading organizations and investigate the opportunities for construction organizations [7]. A three-stage approach (the IMPaKT framework) is presented to develop a business case for KM and the framework could significantly facilitate the implementation of a KM strategy in construction organizations [8]. The survey is made and identified that face-to-face meeting, e-mail and technical business gatherings are perceived as the base transfer mechanisms on knowledge transfer [9]. Udaipurwala and Russell [10] develops intelligent representation structures to store and access construction domain knowledge and couple them with advanced planning tools to enable the quick formulation and assessment of initial construction project plans. EI-Diraby and Kashif [11] proposes distributed ontology architecture developed using rigorous knowledge acquisition and ontology development techniques for knowledge management in highway construction. Although those studies focus on the application of knowledge management in construction, they do not address where to place acquired knowledge for users to find easily, or relationships among knowledge. This study proposes the Network Knowledge Maps (NKM) for construction projects, based on the characteristics of construction project managements. Users are allowed to review available and missing knowledge in core project areas and manage tacit and explicit knowledge appropriately using network knowledge maps. Additionally, the construction Map-based Knowledge Management (MBKM) system is developed for contractors in the construction phase based on network knowledge maps concepts and case studies in Taiwan. 2. Problem statement Construction projects are characterized by their complexity, diversity and non-standard production methods [12]. Professional competency in project management is achieved by integrating knowledge obtained during training with skills developed through experience, and then applying the acquired knowledge in the similar projects [13]. Records of each project, whether successful or unsuccessful, should be kept to identify best and worst company practices. Construction management can be improved by sharing experiences among engineers, helping to avoid mistakes from previous projects. Problems that have already been solved do not need to be solved again. Moreover, engineers and experts normally take domain knowledge with them, leaving little or nothing that will benefit subsequent projects or the company when they finish projects or leave the company. From the perspective of knowledge management, the know-how and experience of construction engineers and experts are the most valuable because its accumulation depends not only on manpower but also on money and time. When knowledge management is considered and utilized in construction industries, general commerce packages currently

focus on document management and knowledge classification. Some construction companies have applied the commerce package for knowledge management. However, those systems can only handle explicit knowledge. Tacit knowledge and experience still exists in the heads of engineers and experts even though the commerce packages have been used in the construction companies. Furthermore, knowledge storage methods must ensure that users can easily find acquired knowledge and illustrate knowledge relationships. The proposed approach is designed to solve the above problems in construction. In the case study, the contractor has specific experience of over 7 years in high-tech building projects. However, high-tech building projects are more risky than other construction projects due to the characteristics of high-tech building projects. The challenge to the enterprise is how to reuse past knowledge and experiences to shorten the project and reduce overhead costs in current and future similar problems. The contractor utilizes the knowledge management integrated with knowledge map to solve the problem and enhance the operation performance. 3. Research objectives Knowledge management in the construction phase mainly deals with creating value from construction operations, organization to company knowledge. Applying and reusing previously finished projects for similar projects in the future is the main issue of knowledge management in the project construction phase. Most engineers and experts surveyed for this study agree that KM is necessary and expect that KM has many advantages in enterprise asset management. However, problems in knowledge management still exist, although the construction enterprises utilize commercial knowledge management tools and software. One problem is that users do not easily find project-related knowledge or know what accumulated knowledge is available. This situation is a major problem for enterprises since most users do not know what and which past project-related knowledge is stored in the enterprise. To solve this problem, this study proposes a Network Knowledge Map to enable engineers and experts to find past project-related knowledge and experience easily. The proposed knowledge map approach is a new concept and approach to improve knowledge management in construction projects. Furthermore, the concept of the knowledge map comes mainly from the practice of construction management in schedule. Previously, most engineers used the network schedule applied in construction management. This study specifically develops a Map-based Knowledge Management (MBKM) system in the construction knowledge management according to the concept and prototype of the network knowledge map. The MBKM system is integrated with the network knowledge map to provide efficient knowledge exchange and management services in construction projects for the reuse of domain knowledge and experience in future related projects. A competitive enterprise must innovatively

Y.-C. Lin et al. / Automation in Construction 15 (2006) 693 – 705

PROBLEM HAPPENING

Problems happen during the construction phase of project

1

I know how to do

How to do

Problem

695

RECORD KNOWLEDGE

CREATE KNOWLEDGE

COLLECT KNOWLEDGE

Explicit knowledge and tacit knowledge are created in solving problems

Explicit knowledge and tacit knowledge are collected in the team groups and enterprise

Explicit knowledge and tacit knowledge are saved and managed in the exchange platform

3

4

2

Knowledge Enterprise Asset

KNOWLEDGE STORAGE

Explicit knowledge and tacit knowledge are maintained as enterprise asset

Explicit knowledge and tacit knowledge are centralized in a system database.

Create New Knowledge Where is knowledge?

Reuse Knowledge New Knowledge Knowledge

10 KNOWLEDGE STORAGE

CREATE NEW KNOWLEDGE

New explicit knowledge and tacit knowledge are updated and centralized in a system database.

New explicit knowledge and tacit knowledge are created in the projects and enterprise

9

8

KNOWLEDGE REUSE

FIND KNOWLEDGE

Explicit knowledge and tacit knowledge are reused in the other or similar projects.

5

Users Utilize knowledge map to fine knowledge they need in the system

7

6

Fig. 1. Knowledge management applied in the construction phase of AEC projects.

utilize knowledge created and accumulated through past completed projects, and share and apply similar knowledge across other related projects. Engineers and experts participating in projects act as knowledge workers facilitating the collection and management of knowledge between current and past projects. To enhance knowledge usage, a knowledge map assist the user as a tool to note key concepts quickly, identify important processes and tools and gain insights into associated behaviors. Fig. 1 shows the main concepts of knowledge management applied to the construction phase of AEC projects. By appropriate modification, the MBKM system can be used in construction companies to support knowledge management functionality for any construction project. 4. Knowledge management in construction projects Construction knowledge management promotes an integrated approach to creating, capturing, accessing, and utilizing a profession’s domain knowledge on products, services and processes. Most knowledge content in the construction phase of a project can be classified as either tacit or explicit knowledge. Tacit knowledge refers to personal, contextspecific knowledge that is difficult to formalize, record or articulate, and is stored in the heads of people [14]. Tacit knowledge also refers to personal knowledge embedded in individual experience and shared and exchanged through direct, face-to-face contact [15]. Moreover, tacit knowledge is difficult to express, represent and communicate [16]. Con-

versely, explicit knowledge is indirectly acquired—it is decoded and re-coded into a person’s mental model, and is then internalized as tacit knowledge. Explicit knowledge can be found in the documents of organizations, including reports, articles, manuals, patents, pictures, images, video, audio and software. Most project-related problems, solution, experience and know-how are in the heads of individual engineers and experts during the construction phase. Implicit knowledge is normally not documented or stored in a system database. Capturing the implicit knowledge and make it in form of explicit knowledge is important for executing knowledge management to preserve implicit knowledge as corporate property. The construction phase of any project creates experience, problem-solving, know-how, know-what and innovation. Knowledge management enables tacit knowledge to be reused for other projects, and accelerates the improvement of operations in the construction phase. Lin and Tserng [17] identified five phases in the construction knowledge management life cycle—knowledge acquisition, knowledge extraction, knowledge storage, knowledge sharing and knowledge update. Each phase is briefly outlined as follows. Most information and knowledge are acquired in the knowledge acquisition phase on the job site in the construction phase. Hence, the accumulation of knowledge on a job site plays a significant role in the knowledge acquisition phase. Most knowledge acquisition work is performed in the office because all information and tacit knowledge sent back

696

Y.-C. Lin et al. / Automation in Construction 15 (2006) 693 – 705

from the job site can be transferred to explicit knowledge. The expert’s thoughts and experience are captured in the knowledge extraction phase. Some knowledge that must be extracted for reuse and storage may only exist in the memories of experts and engineers. Knowledge extraction may also be extended to include capturing knowledge from other sources such as problem-solution descriptions, construction operation process digital records, virtual communication and collaboration. Moreover, knowledge workers assist experts and engineers and deal with the digital process recording work if they are important or valuable as company assets. In the knowledge storage phase, all information and knowledge are centralized and stored in the central database to prevent the collection of redundant data. All information and knowledge can be stored in the system by ensuring that all data are in the appropriate electronic file format. Knowledge sharing refers to the ultimate objective of knowledge management. After the developing knowledge management, individuals who need to apply knowledge of a particular project can access relevant knowledge for reuse. Available knowledge and experience is continually updated in the knowledge update phase. Reused experience can be evaluated in the context of a new problem to be solved, either in terms of the suitability of the selected experience, or in terms of the accuracy of the retrieved experience. Such evaluation is important to ensure that the process of reusing experience continually improves. Invalid knowledge must be identified and eliminated or updated. 5. Methodology—network knowledge map A knowledge map is defined as a diagrammatic and graphic representation of knowledge illustrating how knowledge and knowledge attributes are related. Knowledge map is a consciously designed communication medium using graphical presentation of text, models, numbers or symbols between makers and users [18]. The knowledge map is intended to help users find needed knowledge easily and effectively. Dynamic knowledge map can assist in the reuse of experts’ tacit knowledge (Woo, 2004). Utilizing knowledge mapping has two advantages. First, the knowledge map is represented in a simple, clear visual presentation in the knowledge management system. Second, the mapping methodology helps users to identify key the most critical and easily available knowledge areas of the project. A knowledge map includes the sources, flows, constraints, and sinks (losses or stopping points) of knowledge within an organization [19]. Conventionally adopted knowledge maps help users to identify intellectual capital, socialize new members and improve organizational learning [18]. Knowledge maps have been used for various applications, even including knowledge management software tools [20]. Davenport and Prusak [21] have observed that developing a knowledge map involves locating important knowledge in the organization and then publishing a list or picture showing where to find it. The knowledge map plays significant roles in implementing knowledge management. All captured knowledge can be

summarized and abstracted via the knowledge map. The knowledge map also gives a useful blueprint for implementing a knowledge management system. A research methodology has been proposed and applied in the case studies to improve knowledge management in construction. The proposed network knowledge map is specifically designed for knowledge management in the construction industry. Although knowledge maps not new in KM, the proposed network knowledge maps are novel and specifically designed for construction knowledge management. The components and procedures of knowledge maps are designed based on the practice of construction project management, and are different from other existing knowledge maps. The components and procedures of knowledge maps are described below. 5.1. The components of network knowledge map The network knowledge map proposed in this study consists of three components (see Fig. 2): 1. Knowledge Map Unit: Graphical representation of knowledge with nodes, sub-nodes and linkages: ˝ Node:Rectangular object (representing project or map-unit representation), ˝ Sub-node:Ellipsoid object (representing captured knowledge), ˝ Linkage:Arrow between nodes indicating relationships among knowledge; and 2. Knowledge attribute: Descriptive representation of knowledge feature; 3. Knowledge packages: Extra files to illustrate Knowledge preparation. Like project scheduling management, knowledge management is based on the concept of project planning and control activities. This study proposes a new and original approach, network-based knowledge maps, to improve knowledge sharing and reuse. Fig. 3 gives an overview and conceptual framework of Network-based knowledge maps utilized in construction knowledge management. Knowledge and information linked with knowledge map units in previous projects may be reused and applied in future projects. Information and domain knowledge from all projects are divided and saved as map units in categories related to the projects for collection and management. The main benefit of map-based knowledge management is the ease of understanding and reapplying the information and knowledge. Knowledge is saved in project map units that include both tacit and explicit knowledge. In terms of explicit knowledge, project-related information or knowledge generally includes specification/contract, reports, drawing, change order and data. By contrast, tacit knowledge may include process records, problems faced, problems solved, expert suggestions, knowhow, innovation and experience notes. The information and knowledge is best saved as map units, which makes user classification and searching easy. Additionally, users may

Y.-C. Lin et al. / Automation in Construction 15 (2006) 693 – 705

697

Procure formwork

Mobilize

Forms and Concrete

Excavate footings

Lay out footings

Erect steel

Procure steel

Knowledge Packages

Knowledge Attribute Knowledge ID

Den-D34-K2342

Project

Project 532A

Activity

Erect Steel

Author

Tony W ang

Submit Date

01/12/2002

Approve Date

01/15/2002

Keyword Cite Rate

Erect Steel B Smartsteel ` 32

Similarity Links

23

Edition

3.1

Type

Know how & Know-what service

Format

text

Size

32 m

Attachments

3-1.jpg

Approver Description

3-2.jpg 3-3.mpg 3-4.mpg ` ` ` Jack Lin The new method of erecting steel is proposed for the project 345c. The result of applying new method approves time and cost saving. The name "Smartsteel" is used for the new method. A detailed description is illustrated in the attaching files.

Knowledge Map Unit Fig. 2. Knowledge sharing and reuse using network knowledge map.

search and refer to related information and knowledge from related map units in past projects (see Fig. 4). The knowledge map units and relationships of map-based knowledge management is the same as the activities and relationships of construction project management. Exactly how current and past map units are related helps users link available and related information and knowledge regard to the selected project. Experts or engineers who enter related information and knowledge into the system need to add the relationships concerning the map units in the project map. The system is naturally designed to link among the same or similar knowledge map units together automatically or manually based on high similarity. 5.2. Procedural usage for network knowledge maps Procedures are presented for constructing the network knowledge maps according to the knowledge management

framework. The procedure consists of the following phases: knowledge determination, knowledge extraction, knowledge attribute, knowledge linking and knowledge validation (see Fig. 5). 5.3. Knowledge determination phase This phase encompasses defining knowledge and baseline taxonomy within a project. The scope of the knowledge map determines whether the knowledge map is constructed throughout a specific project. The detailed level of knowledge analysis is determined according to the scope. The appropriate level of detail to satisfy project-based knowledge demand effectively must be determined. Construction project knowledge can be analyzed using map-units when analyzing the source of knowledge within a project. Related knowledge is analyzed based on a map-unit of the construction project. Restated, the map-unit of the selected project determines which

698

Y.-C. Lin et al. / Automation in Construction 15 (2006) 693 – 705 Knowledge Sharing

Other Projects

Knowledge Reuse

Knowledge Asset

Knowledge Acquisition Knowledge Map

Map unit

Similar Knowledge Knowledge

Activity

Project

Knowledge Attribute

Knowledge

Project

Knowledge

Title Author Activity Project Edition Submit Date Approve Date Contents

Activity

Project Map

Activity Activity

Activity Activity

Map Unit

Legend

Knowledge Input Flow

Activity

Knowledge Out Flow

Fig. 3. The application of network knowledge map with knowledge management.

experience and know-how should and can be captured. The types of tacit and explicit knowledge specific to the map-unit are then considered. Moreover, all capturing and documenting knowledge regards to the map-unit are saved in this map category.

include the keywords, description, project name, map-unit name, contributor, and attached files. 5.6. Knowledge linking phase

Two types of knowledge, tacit knowledge and explicit knowledge, are extracted from the projects. Tacit knowledge and explicit knowledge may arise in any project. After identifying content of knowledge from project, the knowledge to be extracted is determined via various technologies. Recommended knowledge extraction techniques include interviewing with experts, group meeting discussion and digital process record.

Identified after completing the knowledge attribute, the knowledge link is initially indicated when the tacit or explicit knowledge is available and documenting, and is later confirmed. This study proposes three types of knowledge linking—map-unit linked to map-unit based on high similarity, map-unit linked to knowledge based on the relationship between map-unit and knowledge, and knowledge linked to knowledge based on high similarity. When the contributor generates a new link, the link has to be examined and confirmed before the knowledge map is published.

5.5. Knowledge attribute phase

5.7. Knowledge validation phase

Knowledge attributes display the basic description of extracted knowledge and derive relationships with project and similarity map-unit. Knowledge attributes largely focuses on providing the basic and required knowledge information for knowledge workers and general users. Knowledge attributes

All knowledge maps have to be validated before publishing the map. All the validation process must be communicated with domain experts, senior engineers, knowledge workers and knowledge map creators in the enterprise knowledge management division.

5.4. Knowledge extraction phase

Y.-C. Lin et al. / Automation in Construction 15 (2006) 693 – 705

699

supports four distinct layers, interface, access, application and database, each with its own responsibilities. The interface layer defines the administrative and enduser interfaces. The users can access information through

6. The system This section describes the MBKM system in detail. Fig. 6 shows the system architecture. The MBKM system server

Knowledge Acquisition and Extraction Phases Explicit Knowledge

Storage

Index

Transfer

Summit

Tacit Knowledge

Edit Comments

Package Summit

Experiences

Enclose

Videos

Documents Photos

Knowledge Package

Knowledge Map

Files

Approve

Publish Knowledge Map

Knowledge Storage Phase

Knowledge System

Update Knowledge

Search

Reuse Knowledge

Knowledge Map

Find Knowledge

Junior engineers Experts

Workers

Senior engineers

Knowledge Sharing Phase Fig. 4. Concept and framework of map-based knowledge management.

Knowledge Update Phase

Knowledge Management Team

700

Y.-C. Lin et al. / Automation in Construction 15 (2006) 693 – 705

Whether the knowledge map is relationship to others

Whether the knowledge is valuable to be maintained

Justify

No

Determine no knowledge to be maintained

Justify

Yes

Only knowledge unit itself without any links

Yes

Link knowledge unit with other units from others similar projects

Yes

Link knowledge unit with other units in the same project

Yes

Step1 Knowledge Determinations Phase

No

Step4 Determine the detail level of knowledge analysis

Determine position of knowledge

Check Vertical Level

Knowledge Linking Phase

No

Select and decide title of knowledge

Check Horizontal Level

Determine range of knowledge

No

Check and Understand where knowledge is

Check explicit knowledge

Whether the knowledge map is suitable to be published

Yes

Collect and manage explicit knowledge regard to knowledge

Validate

NO

Notice to revised and resubmit knowledge map

No Step2 Knowledge Extraction Phase

OK Check tacit knowledge

Yes

Transfer tacit knowledge into explicit knowledge

Yes

Step5 Knowledge Validation Phase

Process knowledge map to be published in the system

Determine to resubmit NO

End of this Process

Package all digitized information (explicit knowledge)

Digitize all explicit knowledge regard to knowledge

Analysis the knowledge Attribute

Note and describe knowledge background

Select knowledge experts in the enterprise

Determine keyword of knowledge

Save and submit to next phase

Select type of knowledge

End of Process

Reject the knowledge map

Step3 Knowledge Attribute Phase

Fig. 5. The procedures of knowledge map implementation.

web browsers such as Microsoft Internet Explorer or Netscape Navigator. Administrators can control and manage information via either the web browser or a separate server interface. The access layer provides system security and restricted access, firewall services and system administration functions. The application layer defines various applications for accumulating and managing information. These applications offer indexing, full text search, collaborative work and document management functions. The database layer consists of primary and backup SQL Server 2003 databases. All project information and knowledge in the MBKM system is centralized in a system database. Project participants may have access to all or some of these documents

through the Internet, as determined by their access permissions. Information or knowledge about the project can be obtained from and deposited into the system database only through a secure interface. The web and database servers are distributed on different computers, between which a firewall can be constructed to protect the system database against intrusion. The system utilizes the network knowledge map approach to represent the locations of explicit and tacit knowledge within a construction project. The network knowledge map in the system provides users with a robust foundation to capture, share and update project-related or map-unit-related knowledge. The benefits of constructing a knowledge map in the system are as follows: (1) efficient capture of knowledge; (2)

Y.-C. Lin et al. / Automation in Construction 15 (2006) 693 – 705

Knowledge Management Team

701

Knowledge Sharing

Experts

Knowledge workers Engineers

Junior Engineers

Knowledge committees

Internet

E-Learning

Construction Map-based Knowledge System Internet Experts

On-site Engineers On-site Engineers Engineers

Knowledge Acquisition

Knowledge Reuse

System Architecture

Java Script Client Personal computer& Web browser (IE)

Interface layer

Users

Security

Authentication

Access layer Knowledge Map JSP

Application Server MS Windows 2003 Server Apache Server Java 2 SDK

Application layer

Engine

Components

E-Course Service Specification/Contract Service

Backup

Database layer

Problem-Solving Service Database Server Video/Photo Service MS SQL Server 2003

Function Service

Document/Report Service E-Meeting Service

Fig. 6. System architecture.

understanding of relationships between knowledge, and (3) formalization of all knowledge inventories within a construction project. The proposed system supports three search functions, by project category, keyword and expert category. The keyword search and project category functions enable users to find knowledge directly from the knowledge map units of selected

projects. Additionally, the system provides another service for users to find the related the domain experts based on required knowledge. The MBKM portal services described in this study are made available to all participants of the corporation through a portal, which also serves as a messaging (mail) server for the enterprise. The portal is a key element of the proposed system

702

Y.-C. Lin et al. / Automation in Construction 15 (2006) 693 – 705

Table 1 Scenario description for using MBKM system Step

Phase

Scenario

Knowledge acquisition

1. Collect information

Knowledge worker collects all activities-related documents/in formation/data Knowledge worker digitizes paper-based information/ documents into the digital in formation/data set Senior engineer edits the description and comment/note for digital document/ information Senior engineer packages the description and comment/note with attaching related files that can illustrate the explanation or example Senior engineer submits pack age that includes the description and comment/note with attaching related files for approving Another knowledge worker assist junior engineer to record digital related process information for the operation of successful and failure events Junior engineer and a knowledge worker edits the description and note/comment for the records of video and photograph Knowledge workers collect and manage the coordinated information (includes grouping the meeting records and knowledge communities) Junior engineer or a knowledge worker packages the description and comment/note with attaching related files Junior engineer or a knowledge worker submits the package that includes the description and comment/note with attaching related files Knowledge worker/expert checks and approves the submission of knowledge package before the classification and storage Knowledge worker classifies the approved knowledge and classified this knowledge by placing it in an appropriate position (map units of project map) in the system Knowledge worker stores knowledge package into system database based on the classification. Knowledge package is automatically backed-up from the system database to another database for the safety purposes.

2. Digital information

3. Edit information

4. Package information

5. Submit information

Knowledge extraction

6. Record operation and event

7. Edit knowledge

8. Manage knowledge

9. Package knowledge

10. Submit knowledge

Knowledge storage

11. Approve knowledge

12. Classify knowledge

13. Store knowledge

14. Backup knowledge

Table 1 (continued) Step

Phase

Scenario

Knowledge storage

15. Publish knowledge

Knowledge sharing

16. Search knowledge

Knowledge package is published after knowledge map is validated and announced for re-use and application New junior engineer found past related knowledge/ experience by using knowledge map and domain expert search New junior engineer refers and studies past existing knowledge/ experience that is stored in the knowledge map units of project New junior engineer modifies the original knowledge package based on new projects or others existing projects New junior engineer applies the modified existing knowledge package in other existing projects or future projects New junior engineer collects feedback from the applied-original or modified knowledge package New junior engineer collects all paper-based and electronic format of document/information/data New junior engineer edits the digital document/information by adding detail description and comment/note New junior engineer packages the description and comment/note with attaching related files that can illustrate the explanation or example Knowledge package would be approved to be processed under an accurate procedure before saving in the system Knowledge package is republished in the project map and announced for the reuse or application, based on the original knowledge

17. Refer Knowledge

18. Modify knowledge

19. Apply knowledge

20. Collect feedback Knowledge update

21. Collect information 22. Remote knowledge

23. Repackage knowledge

24. Approve knowledge

25. Republish knowledge

and consists of three content areas, public, member, and knowledge manager. The public area is available to anyone who is interested in the project. The project member and knowledge manager are accessible only to members with passwords. Project members can use MBKM services in the project member area, including messaging, data access, and files access. Additionally, knowledge managers can log into the manager area, where they can access all information in the project server. 7. Case study In the following case study, the contractor with 7 years of experience involving six High-Tech factory building projects decided to apply knowledge management to raise the enterprise competition. The contractor hoped to maintain knowledge and experience effectively from the senior engineers and experts specific in the construction phase,

Y.-C. Lin et al. / Automation in Construction 15 (2006) 693 – 705

and so decided to hire two High-Tech construction knowledge workers, to help the senior engineers manage project execution knowledge and experience gained from ten previous finished projects. To reuse knowledge in future similar projects, the company decided to exploit knowledge management to pass on the valuable know-how to the

703

engineers and manage it well to keep the knowledge inside the company. The contractor applied and practiced knowledge management using a network knowledge map. Additionally, Table 1 presents the details of various stages of construction knowledge management implementations. Fig. 7 illustrates the system utilized in this case study.

Current Project

Past Projects

Project

Construction Phase

Knowledge Asset

Experts and Engineers

Project Knowledge

Explicit Knowledge

Knowledge Map

Knowledge Management Teams

Knowledge Map

Map-based Knowledge Management

Knowledge & Experience Sharing Engineers

Manager

Knowledge Management Team

On-site Engineer Head Office

Fig. 7. The application of knowledge sharing in MBKM system.

704

Y.-C. Lin et al. / Automation in Construction 15 (2006) 693 – 705

A knowledge worker collects and digitizes information/ documentation from the current project and six finished projects. The senior engineer then edits the descriptions and notes and packages them as knowledge for submission.

knowledge maps. He begins to reuse knowledge from the previous seven finished projects, and applies the knowledge to his own new project. Additionally, the junior engineer gives experienced feedback and offers knowledge that can be reused in future problems. Another engineer also later reuses the same knowledge to solve the same problem in another new project.

7.2. Knowledge extraction phase (steps 6 –10)

7.5. Knowledge update phase (steps 21 – 25)

A junior engineer and another knowledge worker record all the operating procedures by taking digital video and photographs in the executed project. The senior engineer discusses progress with the expert every 2 days to accelerate problem solving. All discussions were recorded and summarized as recommended by the senior engineer. Discussions with experts continue for 6 months, until the problem is solved. Most engineers are expected to provide their own knowledge on their assigned tasks. The junior and senior engineers record and summarize their experiences and domain knowledge in the system, so that the solutions can be reused in future projects. The domain knowledge includes the problem description (including documents, photographs, drawings and specifications), the solution (including related documents and photographs and video of processes) and expert suggestions (such as notes, discussions and meeting records). Finally, items of domain knowledge and experience are linked to map units.

The junior engineer solves his problem in collaboration with senior engineers using knowledge obtained from previous similar projects. Finally, the junior engineer notes and submits the new suggestion and experience in the project map units, linked with the original knowledge. Furthermore, the knowledge is updated with the additional feedback and solution. The updated knowledge set is then republished in the map units of the project after the approval process is completed, and the notice message is transmitted to the authorized members.

7.1. Knowledge acquisition phase (steps 1– 5)

7.3. Knowledge storage phase (steps 11– 15) When the submitted knowledge set is approved, a knowledge worker in the knowledge management team attributes knowledge and places it in an appropriate position (according to project map unit) in the system. In other words, users can find and read related domain knowledge directly by simply clicking on project map units. All knowledge maps have to be validated to perform well before the map is published. Domain experts, knowledge worker, and knowledge map creators all help perform validation in the enterprise knowledge management team. Finally, the knowledge set is automatically backed up from the knowledge database to another database. After approving and storing knowledge, the system transmits a message to the appropriate users automatically stating that the knowledge has been updated. 7.4. Knowledge sharing phase (steps 16 – 20) A new project is started after High-Tech building project was constructed 10 months previously. Another junior engineer with no prior experience encounters a similar problem special in Fire alarm system and attempts to solve it by finding past knowledge/information. The junior engineer utilizes the knowledge map search to find an expert with domain knowledge concerning High-Tech building project and which knowledge are available according to the knowledge map in seven finished projects. The junior engineer identifies the relevant experts, and then retrieves and studies the knowledge packages (including digital video and documentation) from the

8. System evaluation Verification and validation tests were performed to evaluate the system. Verification aims determines whether the system operates correctly and behaves according to the design and specification, while validation assesses the usefulness of the system. The verification test was performed by verifying whether the MBKM system can perform tasks as specified in the system analysis and design. Selected projects participants were asked to use the MBKM system, and then provide feedback by answering questionnaires. The case projects participants included: five project managers with 5 years of experience; five senior engineers with 20 years of experience; three senior engineers with 10 years of experience, two junior engineers within 1 year of experience, Table 2 System evaluation result Mean score The functionality of system Ease of acquiring experience/knowledge Comprehensive Accuracy Reliability Applicable to Construction Industry

1.8 1.7 1.6 1.4 1.8

The use of system Ease of use User interface Over information sufficiency Over system usefulness

1.8 1.4 1.5 1.7

The capability of system Reduce unnecessary time Reduce unnecessary costs Reduce problems mistakes Improve knowledge search Improve knowledge sharing Improve e-learning performance

1.4 1.5 1.7 1.8 1.6 1.5

The mean score is calculated from respondents’ feedback on five scale questionnaire ( 2 (not useful), 1, 0, +1 and +2 (very useful)).

Y.-C. Lin et al. / Automation in Construction 15 (2006) 693 – 705

and two knowledge workers within 3 year of experience. Table 2 illustrates the system testing results. 9. Conclusions Internet-based technologies and knowledge management concepts can be effectively utilized during the construction phase of a project, to enable knowledge to be captured and reused in similar projects. This study proposes using a network knowledge map to apply knowledge management in the construction phase of construction projects. Map-based knowledge management (MBKM) and a knowledge-sharing platform for construction projects are presented. The construction mapbased knowledge management system maps organize valuable information and knowledge into map units during the construction phase of a project. The MBKM system provides insight into the factors affecting construction management, and so helps engineers share knowledge and improve the results of the entire construction project. Junior engineers can interact with the computer to obtain domain knowledge, and thus prepare for and participate in a construction project. In summary, the MBKM system can assist engineers by providing structured and unstructured information to support reference to, and the reuse of, explicit and tacit knowledge. Integrating web-based knowledge management and network knowledge maps appears to be a promising mean of enhancing construction operation and management, particularly in the construction phase of a project. The case study of the new High-Tech factory building located in Taiwan demonstrates that the MBKM system effectively promotes the sharing of knowledge acquired from past six projects and reuse for other similar projects. The case study also highlights the need to improve knowledge exchange and management platforms. However, users gave the following feedback on the use of the system: (1) the past experience and knowledge in the knowledge management team is difficult to acquire and manage using the knowledge database content; (2) most senior engineers and experts are not willing to share their knowledge and experience without a proper reward policy and strategy; (3) editing and recording knowledge without any assistance from knowledge workers is time-consuming and inconvenient for senior engineers, and (4) most senior engineers agree the MBKM system is a helpful platform for editing and managing knowledge and experience. Network knowledge maps are mainly utilized to help find required knowledge easily and effectively. According to the questionnaire responses, the primary benefits of utilizing a network knowledge map are as follows: (1) the knowledge map clearly and visually identifies key knowledge areas that are most strategic and critical to the project, and (2) deals with the assistance for users to find the needed knowledge easily and effectively. Acknowledgement The authors would like to acknowledge the National Science Council, Taiwan, for financially supporting this work under contract number NSC-2211-E-231-001 and express our

705

appreciation to the G-H Construction Inc. for assistance in the system design and interviews, and to the experts and engineers of the owner of the project for providing useful data, valuable information, and helpful comments during system design and development. References [1] Y. Malhotra, Knowledge Management and Virtual Organizations, Idea Group Publishing, Hershey, PA, 2000. [2] Y. Malhotra, Knowledge Management and Business Model Innovation, Idea Group Publishing, Hershey, PA, 2001. [3] Alberto Carneiro, The role of intelligent resources in knowledge management, Journal of Knowledge Management 5 (4) (2001) 358 – 367. [4] Ralph Bergmann, Experience Management: Foundations, Development Methodology, and Internet-Based Applications, Springer, Germany, 2002. [5] J.M. Kamara, G. Augenbroe, C.J. Anumba, P.M. Carrillo, Knowledge Management in the architecture, engineering and construction industry, Construction Innovation 2 (2002) 53 – 67. [6] Jeong-Han Woo, Mark J. Clayton, Robert E. Johnson, Benito E. Flores, Christopher Ellis, Dynamic Knowledge Map: reusing experts’ tacit knowledge in the AEC industry, Journal of Automation in Construction 13 (2004) 203 – 207. [7] Patricia Carneiro, Managing knowledge: lessons from the oil and gas sector, Journal of Construction Management and Economics 22 (6) (2004) 631 – 642. [8] H.S. Robinson, P.M. Carrillo, C.J. Anumba, A.M. AI-Ghassani, Developing a business case for knowledge management: the IMPaKT approach, Journal of Construction Management and Economics 22 (7) (2004) 733 – 743. [9] B. Jan, R. Sara, Waara Fredrik, Cross-border post-acquisition knowledge transfer among construction consultants, Construction Management & Economics 22 (4) (2004) 421 – 427. [10] A. Udaipurwala, A.D. Russell, Computer-assisted construction methods knowledge management and selection, Canadian Journal of Civil Engineering 29 (3) (2002) 499 – 516. [11] T.E. EI-Diraby, K.F. Kashif, Distribute Ontology Architecture for Knowledge Management in Highway Construction, Journal of Construction Engineering and Management 131 (5) (2005) 591 – 603. [12] Richard H. Clough, Clenn A. Sears, S. Keoki Sears, Construction Project Management, 4th edR, Wiley, New York, 2000. [13] F.T. Edum-Fotwe, R. McCaffer, Developing project management competency: perspectives from the construction industry, International Journal of Project Management 18 (2000) 111 – 124. [14] A. Hart, Knowledge Acquisition for Expert Systems, 2nd edR, McGrawHill, New York, 1992. [15] A. Tiwana, The Knowledge Management Toolkit-Practical Techniques for Building a Knowledge Management System, Prentice-Hall, New Jersey, 2000. [16] I. Nonaka, H. Takeuchi, The Knowledge-Creating Company, Oxford University Press, Oxford, 1995. [17] Yu-Cheng Lin, H. Ping Tserng, Knowledge management and its application to lean construction, Proc. Eleventh Annual Conference of the International Group for Lean Construction (IGLC-11), July. 21 – 25 2003, held in Virginia, USA, 2003, pp. 542 – 552. [18] M. Wexler, The who, what, and why of knowledge mapping, Journal of Knowledge Management 5 (3) (2001) 249 – 263. [19] J. Liebowitz, Linking social network analysis with the analytic hierarchy process for knowledge mapping in organizations, Journal of Knowledge Management 9 (1) (2005) 76 – 86. [20] M. Noll, D. Frohlich, E. Schiebel, Knowledge maps of knowledge management tools: information visualization with BibTechMon, in: D. Karagiannis, U. Reimer (Eds.), Practical Applications of Knowledge Management 2002 Conference Proceedings, Springer-Verlag, New York, NY, 2002. [21] T.H. Davenport, L. Prusak, Working Knowledge, Harvard Business School Press, 1998.