Automation in Construction 23 (2012) 7186

Contents lists available at SciVerse ScienceDirect

Automation in Construction
journal homepage: www.elsevier.com/locate/autcon

A framework for facilitating multi-dimensional information integration, management and visualization in engineering projects
I-Chen Wu a, Shang-Hsien Hsieh b,, 1
a b

Department of Civil Engineering, National Kaohsiung University of Applied Sciences, Kaohsiung, Taiwan Department of Civil Engineering, National Taiwan University, Taipei, Taiwan

a r t i c l e

i n f o

a b s t r a c t
This research proposes an application framework, named PIIM Framework (Project Information Integration Management Framework), to effectively solve the problems of integrating project information and system interfaces among different participating parties and engineering application systems, as well as to provide ve kernel modules to encapsulate complicated management and visualization functions. The PIIM Framework is constructed on the basis of object-oriented techniques, such that users can take advantage of object oriented programming to easily develop applications required. In addition, this research conducted an actual engineering project example, and developed a prototype system based on the PIIM Framework to verify its feasibility. Through the feasibility study, it has been demonstrated that the PIIM Framework can provide not only provide a more efcient method to integrate, manage and visualize project information, but can also save programming time for the project management team when developing a project management system from scratch. 2011 Elsevier B.V. All rights reserved.

Article history: Accepted 12 December 2011 Available online 20 February 2012 Keywords: Project management nD 4D Information integration Information visualization

1. Introduction Today's construction projects are becoming ever more complex and time driven, especially as the amount of project data and active project participants increases. Several known issues pertaining to this phenomenon are described as follows. Firstly, the construction industry continually seeks to apply new information technology (IT) based methods to substitute the traditional paper-based management methods, with the need for information to be created just once and used over the entire project lifecycle, rather than the generation of the same information multiple times in current processes. However, the diverse range of project participants still wish to maintain their own data, structured to suit their own specialized needs. Hence, the differences of data denition, data format, and data storage increase the difculty of integrating the data from diverse sources in engineering projects. Secondly, current information management methods are based on le collection approaches, which have no relationship with each other. Therefore, project managers cannot consider engineering information synthetically in performing project control. If they desire increased accuracy and total control of engineering projects, project managers must spend time collecting and handling data which are generated from different systems. Thirdly, although much of this engineering information is produced electronically and is visual in nature, teams primarily use paper-based views of engineering information to share information with one another in project meetings.
Corresponding author. Tel.: + 886 7 381 4526x5238; fax: + 886 7 383 1371. E-mail addresses: kwu@kuas.edu.tw (I.-C. Wu), shhsieh@ntu.edu.tw (S.-H. Hsieh). 1 Tel.: + 886 2 3366 4313; fax: + 886 2 2368 8213. 0926-5805/$ see front matter 2011 Elsevier B.V. All rights reserved. doi:10.1016/j.autcon.2011.12.010

The project participants would not be able to understand the project status comprehensively due to the lack of an intuitive and userfriendly visualization tool to display the related project information. Fourthly, the participants in the project need to communicate and transfer information with each other, either within a phase or between two phases. However, information exchange between project parties is limited to the digital les, a medium in which retrieval and exchange is inconvenient and inefcient. In addressing the abovementioned issues, the project management system has become increasingly important in a project management effort. The role of the project management system is to provide assistance to the project management team in handling complicated project information, and providing useful and signicant information for controlling the project and supporting decision-making. At present, the project management systems which are employed in the construction industry can be divided into two types. The rst type is commercial software, in which projects are managed using mostly Gantt Charts, Program Evaluation and Review Technique (PERT) [13] and Critical Path Method (CPM) [4]. These management techniques have spread quickly into many private enterprises and represent signicant commercial value. Thus, much of the related software and modules have been developed by commercial software companies, for example, Microsoft Project, Oracle's Primavera P6 Professional Project Management, Daptiv Project Portfolio Management (PPM) and Project Management Module in SAP's ERP solution. The second type of project management systems is in-house software, which is usually utilized where the commercial software does not offer sufcient or special functionalities for use in particular engineering projects or rms. Therefore, some rms will develop the in-house

72

I.-C. Wu, S.-H. Hsieh / Automation in Construction 23 (2012) 7186

software to meet their needs. Examples of this include Bechtel [5], Parsons Brinckerhoff [6,7], Kajima [8,9], CECI, CTCI [10,11], RUENTEX, SINOTECH among many others. While the abovementioned systems can assist project manager in managing the engineering project, they still leave signicant issues unresolved. This may be considered as two issues: (1) Traditional project management systems mainly provide text, graph and complicated network schedules for controlling project and making decisions. Today's complex and time-driven projects, coupled with increasingly large project information volumes and active project participants, require more effective integration, management and communication tools than that which traditional project management systems can provide. Because of the complexity of the construction industry, the multiple phases of the construction project life-cycle combined with project participants using various heterogeneous systems causes information integration to be an important and imperative step in achieving efcient and effective collaboration for the project's success. For example, the construction project team can understand the conicts in relation to time and space in advance by using the conict detection mechanism, where the computer informs team members of parts of the building which are in conict or will clash, by the integrating schedule information and 3D model. In addition, this allows quality control to occur at any time and at any place during the engineering project. The aim of quality control is to oversee for the completion of the engineering project in a timely manner through the most protable and cost-effective way. These questions arise in relation to multi-dimensional information integration, management, and visualization of engineering projects. It follows from what has been discussed above that an effective project management system should provide not only provide sufcient and comprehensive information to facilitate project management, but also the various visualization tools to assist with information distribution and communication; (2) In the construction industry, the contents of project management will be changed to meet special engineering requirements and contract stipulations in every project. Even though commercial project management systems can provide the most commonly required functions, they often do not have enough exibility to change management workows or enhance certain functionalities when required. On the other hand, some construction management rms develop in-house project management system for fullling specic project requirements, at an immense cost of time and money on development and maintenance efforts. In this research, we propose a Project Information Integration Management Framework (PIIIM Framework) to address the aforementioned issues. The PIIM Framework is a reusable and semicomplete application that can be customized to deliver specic applications. We hope that future developers will be able to efciently develop a useful system for construction project management on the basis of the PIIM Framework. We also develop the prototype Visual Project Management Information System (VisPMIS), based on the proposed PIIM Framework, to assist project management teams in resolving the issues of integration, management and visualization of construction project information. Through this example application, the PIIM Framework proposed in this research is demonstrated and the functionality is veried. 2. Related works In this section, we briey review the main literature related to our work, with the purpose of highlighting our contributions in this research area. Construction project management is responsible for the application of modern management techniques and systems to the execution of a project from start to nish and to achieve predetermined objectives of scope, quality, time and cost to the equal satisfaction of those involved. However, development is required on various

related topics under the following focus areas: (1) Because of the complexity of the construction industry, the multiple phases of the construction project life-cycle, and the project participants using various heterogeneous systems, information integration becomes an important and imperative step to achieve efcient and effective collaboration for project success; (2) The use of visualization to present engineering information is an effective and efcient method for the distribution of information and communication with project participants during project meetings. A detailed discussion is presented in the next section. 2.1. Multi-dimensional information integration During the life cycle of an engineering project, a voluminous amount of data and information is usually created along the delivery processes of construction products. The construction project team must consider a wide variety of information when controlling the project and making project decisions. In general, teams primarily use paper-based views of project information and share information through les. Because the information dened in the information distribution systems of various parties are usually different, they are always difcult to integrate with one another for the purpose of information communication and distribution. Important relationships between different pieces of project information are not communicated effectively as construction information is not effectively integrated and used. At present, some data models have been established for integration purposes. The Building Information Model (BIM) [12] is a computer model database of building design information, which may also contain information about the building's construction, management, operations and maintenance. Some research examples include Azhar et al., who demonstrated the use of BIM for sustainable design and the LEED certication process [13], and Hu et al., who applied BIM technology to address conicts and structural safety problems during construction [14,15]. However, as these solutions are based on different and incompatible standards, an open and neutral data format is required to ensure data compatibility across the different applications (e.g. Autodesk Revit, Bentley TriForma, Graphisoft ArchiCAD and so on). Industry Foundation Classes [16], developed by the International Alliance for Interoperability, can provide such integration and interoperability capabilities. The IFC data model is an object-oriented data model based on class denitions representing the items (elements, processes, shapes, and so on) that are used by software applications during a construction or facility management project. Implementations of IFC have been reported in various construction IT system integration projects [1720]. In addition, Salford University proposed the concept of nD Modeling, enabling multidimensional information integration, which it foresees as essential in the future of construction management [21,22]. Although IFC can fulll the goals of data sharing and information integration, it still falls short of providing the related objects and processes for project management. The current research attempts to integrate multidimensional information into the database through programming, or provide related project data on 3D models using CAD software API. This approach, relying on proprietary software, is difcult to reuse without the particular software, and is constrained by xed development methods. In this research, we provide a more exible and simple mechanism to handle the multi-dimensional information integration. Firstly, we design and develop the integrated data model to describe and store all the project information. Therefore, the integrated data model can be used to demonstrate the entire construction lifecycle including project management, the processes of construction, and facility operation. Following this, we provide not only a special module that contains a number of importing functions to address the existing project information into the integrated data model, but also provide a management module to access the integrated data model including functions such as data creation, data query, data

I.-C. Wu, S.-H. Hsieh / Automation in Construction 23 (2012) 7186

73

deletion, data insertion, data update, and so on. Users will be able to employ our modules and the integrated data model to build their application to address the multi-dimensional information integration without any commercial software. 2.2. Visualization for communication and information distribution Treicher [23], an experimental psychologist, conducted a famous psychological experiment to determine how human beings obtain information. Through a large number of experiments, he proved that 83% of the information received by human beings is by the sense of sight, and the remainder is by the senses of hearing, smell, touch and taste, supporting the view that information visualization is the most essential for communication and information distribution. Liston et al. [24,25] studied the problem of communication in engineering workspaces, which could be variably dened as physical or virtual spaces where people work, share and use information. From observations of numerous design and construction review meetings, they discovered that teams spent most of their time on descriptive, explanative, evaluative and predictive tasks. All these tasks are critical to enabling better decision-making by project managers. In this research, we employed the aforementioned four tasks to analyze the current information distribution system in construction projects to determine whether sufcient information was being provided. The information distribution system may be classied into four types according to their characteristic of information visualization. The rst type is the 1D system, which conveys information through text, attributes and tables. Examples of these include project websites, databases and general management information systems. Such systems record and describe the project information in detail. The second type is the 2D system, which presents information mainly in the forms of graphs, charts or maps, such as output from GIS, ERP, 2D CAD, and so on. These systems assist the project manager in explaining project decisions, the rationale of the project schedule, and also support them in evaluating project goals and ensuring that project requirements are met. The third type is the 3D system which uses 3D models to present the actual project status and analyze space conicts, with examples including 3D CAD and virtual reality systems. These applications harness 3D visualization to assist descriptions of engineering project status, and enhance communication and discussions during project meetings. The fourth type is the 4D system which binds 3D models with their corresponding construction schedules in their simplest forms. This has recently emerged as a method of providing engineers with an effective tool for managing the complexity of coordination as well as conicts before the actual commencement of construction. Some examples include 4D CAD and construction simulation systems. The realization of 4D CAD technology has also been greatly accelerated by the availability of powerful commercial 4D CAD tools, such as Bentley's Navigator, Intergraph's SmartPlant Review, BALFOUR's FourDscape, Common Point's Project 4D and ConstructSim [26]. There have been numerous, albeit similar, research studies into the use of 4D systems. The important detailed study by Stanford University was instrumental in initiating interest in this area [2729]. Chau et al. built a 4D-GCPSU for construction management [30,31]. Hsieh et al. developed a Construction Director, a 4D simulation system for plant construction [10,11]. Huang et al. developed a virtual prototyping system for simulating construction processes [32]. Dawood and Sikka [33] provided quantitative evidence that a 4D model can increase the efciency of communication along with the interpretive ability of a construction project team. Even though 4D technology can increase the ability of project managers in providing better management and visualization in construction project management, there is room for further development in providing sufcient information for uncovering problems and making better decisions. Table 1 summarizes the different task emphases and possible outcomes of the four information visualization systems that have been described.

Based on Table 1, we are of the view that an efcient information distribution system ought to support all four kinds of characteristics of information visualization at the same time. Therefore, this research proposes a new concept of information visualization, named the 4D + Multi-Data-View, to present the integrated data of engineering projects, with advances towards providing signicant useful information for use in project management and decision-making. The meaning of 4D + is to provide the additional characteristics of information visualization on top of the base 4D information visualization. Hence, users will be able to view the text which describes the project information, the graph to view the project statistics and progress, the 3D models to present the building status and the 4D models to display the construction simulation in the same view. Furthermore, the MultiData-View implies that users will be able to query and view the integrated data with the different views according to their requirements as shown in Fig. 1. In this approach, users will be able to nd the necessary and relevant data they need in a more direct and accessible way. This research developed a visualization and management module to realize this concept. The management module provides functions to acquire the required data the user needs, and passes it on to the visualization module to display the results in different ways according to the user's requirements. 3. PIIM Framework 3.1. Overview As shown in Fig. 2, the PIIM Framework is an application framework which can be divided into three tiers. The rst tier is the integrated data model, which describes and stores information about entire engineering projects; the second tier is the data access interface, which mainly deals with data access between the integrated data model and kernel module; and the third tier is the kernel module, which provides core project management services and encapsulates complicated project management functions into ve easy-touse modules: DataAdapter, DataManager, DataAnalyst, DataViewer and DataExporter. 3.2. Integrated data model This research proposes an integration data model which can be used not only for storing engineering information, but also providing internal system data in the PIIM Framework. As shown in Fig. 3, the integrated data model can be broken down into three parts. The rst part is the IFC Data Model, which is employed to describe most of objects in engineering project management, for example, 3D objects, schedule objects, cost objects and document objects. The second part is the User-Dened Data Model, which works with the IFC Data Model to extend its insufcient management objects, such as quality statistics objects and scope objects. The third part is made up of Relationship Objects which dene the relationship between IFC objects and User-Dened objects. Both User-Dened Data Model and Relationship objects are described by XML. This integrated data model has the important characteristic of the multi-directional data link, as shown in Fig. 4. The Scope item is parent to a lot of Activity items and Activity item is related to many other items, which can be 3D Object items, Cost items,
Table 1 Characteristics of different information visualization. 1D system Descriptive Explanative Evaluative Predictive 2D system 3D system 4D system

74

I.-C. Wu, S.-H. Hsieh / Automation in Construction 23 (2012) 7186

3D object

Cost Item

Schedule Item Quality Item Document Item

Scheduler

ScheduleView

Data Model

3D object

Schedule Item

Cost Item Quality Item Document Item

Accountant

Cost View
Fig. 1. The concept of Multi-Data-View.

Data Model

Document items, and so on. Each of the items records the related and relating relationship. Thus, users will be able to easily nd the related items through these multi-directional links. For example, a designer would be able to understand the related cost and schedule information through linking from the relevant 3D object items; a planner would be able to view the related 3D model by linking from the relevant Scope, Cost and Schedule items. As Fig. 5 shows, we adopt the XML technique to organize and construct the integrated data model. The XML structure can describe the User-Dened Data Model and Relationship Objects, as well as saving the key attribute of IFC Objects for use as a link for matching and obtaining the data from the IFC Data Model. Establishing the integrated data model provides four benets: (1) it can be used for storing the information of the entire project; (2) it can be employed to decrease errors made. For example, the construction project team will be able to understand in advance the conicts in terms of time and space by using the conict detection mechanism where the computer informs team members about parts

of the building in conict; (3) it can be utilized to provide signicant information for project management and decision-making. For example, a construction project team could conduct the Earned-Value analysis method, which includes schedule and cost data for analyzing and reviewing the relationship between schedules and cost overall; (4) users will be able to nd the data they need efciently and effectively through the relationships of multi-directional links. 3.3. Data access interface The data access interface will increase the ease of posing basic queries and update requests to the integrated data model to which it is connected, regardless of the native language of the integrated data model. It will also allow access to the integrated data model in their native language for more complex requests. As mentioned above, the integrated data model is established on both the IFC Data Model and the XML Data Model. Thus, the PIIM Framework involves two types of data access interface, one being the IFC data access interface and the other being the XML data access interface. In this research, the IFCsvr ActiveX Component [34], an open source library for handling IFC data, is employed. It includes features such as a dictionary of IFC schema information and functionality for representing and manipulating instances of IFC objects. Furthermore, we developed the XML data access interface using Microsoft products and technologies. The System.Xml namespace provides standardsbased support for processing XML, which is available in the .NET Framework. 3.4. Five kernel modules The PIIM Framework allows for the segmenting of complex items into manageable modules according to the particular purpose, and encapsulates complex implementation details behind a stable interface. The modules of PIIM Framework are similar to Lego pieces which can be assembled in many ways to suit different purposes. This approach enables applications to be built quickly, and saves development effort. A more detailed discussion follows. 3.4.1. DataAdapter DataAdapter provides functions for receiving and transforming the common data formats encountered in construction into the PIIM Framework. DataAdapter can support approximately 80% of the data

Fig. 2. The three tiers of PIIM Framework.

I.-C. Wu, S.-H. Hsieh / Automation in Construction 23 (2012) 7186

75

IFC Data Model

User-Defined Data Model

3D Objects

Schedule Objects

Quality Statistics Objects

Scope Objects

Relationship Objects

Cost Objects

Document Objects

Fig. 3. The integrated data model of the PIIM Framework.

formats generated from commercial software packages. This module provides three functional classes. The rst is C3dObjectAdapter, which assists the program to import 3D models created using any commercial CAD software (for example, ArchiCAD, Triforma, and Revit). The second is CScheduleAdapter, which is capable of importing Project XML data which is generated from MS Project or Primavera P3, and then transforming it into the IFC 2 3 data format which is employed in the integrated data model for integration. The third is CCostAdapter, which can adapt and manipulate the Microsoft Ofce Excel data format for cost data integration. 3.4.2. DataManager DataManager is responsible for data management and manipulation. The main functions are divided into four parts, as follows: (a) Objects Management: the DataManager includes many different classes which enables the provision of functions for manipulating objects, such as create, modify, delete and query; (b) Data Structure: the DataManager denes some data structure for use in the PIIM Framework, such as class CPiimTreeNode for viewing multi-dimensional information; (c) Simulation Table: the class CSimulationTable can provide functions for collecting and storing the required data for simulating construction processes, which can decrease data handling time; (d) Others: the DataManager will provide functions for assisting the internal operations

within the PIIM Framework, such as opening and saving the PIIM Project which includes the integrated data and the related views and graphs.

3.4.3. DataAnalyst This research employs the Earned-Value analysis method [3537], which is a project management technique used for measuring the project progress in an objective manner, and for controlling and monitoring the construction project. The EVM function in the VisPMIS combines measurements of schedule performance (i.e., behind/ ahead of schedule) and cost performance (i.e., under/over budget) within a single integrated window. When properly applied, EVM can provide an early warning of performance problems and communicate this to all stakeholders. After the integrated data model is established, the class CCalEarnedValue of DataAnalyst can provide the calculation functions of Earned Value analysis, including for example, Budgeted Cost of Work Schedule (BCWS), Budgeted Cost of Work Performed (BCWP), Actual Cost of Work Performed (ACWP), Schedule Variance (SV), Cost Variance (CV), Cost Performance Index (CPI) and Schedule Performance Index (SPI). DataAnalyst also provides the indispensable calculation functions of the schedule progress and quality statistics.

3D Object

Scope

Activity

Cost

Document

Fig. 4. Multi-directional data link of the integrated data model.

76

I.-C. Wu, S.-H. Hsieh / Automation in Construction 23 (2012) 7186

<<XML Data Model >>

Key Attributes Of IFC Objects

IFC Data Model

Reference

Fig. 5. Example for the data structure of the integrated data model.

3.4.4. DataViewer A good visual representation of data can assist people in efciently acquiring applicable information. To achieve this, DataViewer provides ve styles of visualization for communication among the project participants: (a) 1D Visualization: to show the attributes of objects and general project information; (b) 2D Visualization: to show the important graphs and charts related to the objects and project information; (c) 3D Visualization: to display the 3D models of the building; (d) 4D Visualization: display of the various 3D models incrementally at the various stages of construction activities as construction progresses over time; and (e) nD Visualization: to show the integrated data after data binding. 3.4.5. DataExporter For data exchange and sharing, DataExporter can provide functions for exporting project data to three different le formats. The rst is the IFC data format, which is used for data exchange with IFC-based systems. The second is the ifcxml data format, which is utilized for data exchange with general information management systems using XML technology. The third format is the GML data format, which is used for integrating with 3D-GIS geo-data [38]. 3.5. Summary The PIIM Framework is constructed on the basis of object-oriented techniques. Object-oriented analysis and framework design are characterized by features such as modularity, reusability and extensibility, which are recognized as facilitative to the evolution and the maintenance of application systems.

In terms of modularity, the PIIM Framework segments complex objects into manageable pieces or modules according to particular purposes, and encapsulates complex implementation details behind stable interfaces. The modules are similar to the Lego pieces which can be assembled and connected in many ways, to construct different applications easily. Fig. 6 illustrates how the import function for 3D objects is developed. Users can utilize the functionalities of the DataAdapter for receiving 3D Objects, the DataManager for transforming data structure and the DataViewer for viewing 3D models into the application system. Hence, applications can be built quickly and development effort can be reduced. In terms of reusability, the PIIM Framework enhances reusability by dening generic components that can be reapplied to create new applications. For example, coding is achieved in the same way as calculating the percentage of scheduled progress and quality statistics in engineering projects, and so a class CCalPercentage which calculates the percentage is provided in the DataAnalyst for both to use, as shown in Fig. 7. Lastly, the PIIM Framework is designed to include hook methods and mechanisms for enhancing the system with new capabilities, without requiring major changes to the system infrastructure. Fig. 8 shows an example of extensibility in PIIM Framework. The interface IGetEntityAndAttribute is used to specify the required attributes and behavior of object querying. From the research above, two positive achievements are derived for application development. One is that the same interface can be inherited to implement the different classes. Once the application is based on the PIIM Framework, the developer can add new querying features to an existing application without having to change any existing code. Another is that new attributes and behavior can be added into the interface, which would be

Modularity
frm3D Return

DataAdapter

C3dObjectsAdapter frmMain

DataViewer

DataManager

frm4DView

AxOctagaModellerX

Return

C3dObjectsTreeView

Fig. 6. The modularity of PIIM Framework.

I.-C. Wu, S.-H. Hsieh / Automation in Construction 23 (2012) 7186

77

Reusability
frmOne

DataAnalyst percentage of scheduled progress

Return Return

percentage of quality statistic

frmTwo

CCalPercentage

Fig. 7. The reusability of PIIM Framework.

updated for the related class automatically. In this way, the functions of classes will be effectively extended. 4. Prototype system This research prototyped a visual project management information system for a construction management rm named VisPMIS to test the feasibility of the PIIM Framework. 4.1. Engineering example In this research, we selected an actual engineering project in the campus of National Taiwan University (NTU) as a case study. The project involved the construction of an underground parking area and an information center. As illustrated in Fig. 9, the organization of this project is divided between three main participants and areas of responsibility: (a) NTU, the property owner with contractual relationships to Company A and Company B; (b) Company A, a professional engineering and consulting service provider which provided professional construction management services to NTU, while supervising and managing Company B; and (c) Company B, a construction rm which provided the physical workforce for the design and construction of this engineering project. Each of the major project participants executed their respective project management responsibilities.

However, in this research, the majority of the work was the responsibility of the construction management rm. We analyzed the project management requirement by interviewing Company A and NTU. After interviewing NTU, we were able to understand the responsibilities of Company A as stipulated in the contract and as requested by the owner. After interviewing Company A, we were able to visualize the project management process in a practical context. 4.2. Requirements analysis of prototype system The main objective of Company A was to monitor and control all aspects of a project to achieve project goals, such as completing the project on-time and to the specied cost, quality, and performance. The work performed by Company A was mainly focused on project management during engineering construction. As shown in Fig. 10, the main work activities could be divided into three phases: (1) the Project Preparation Phase, (2) the Project Construction Phase, and (3) the Project Completion Phase. During the Project Preparation Phase, the main focus of Company A before construction was project information collection and verication. Company A obtained the planned project information from Company B and other project participants. The information included schedule information, cost information, 3D models of the project, and relevant documents. After receiving the project information,

Extensibility
DataManager
<<Interface>>

frmQuery Return

IGetEntityAndAttribute

CGetEntityAndAttributeByName

CGetEntityAndAttributeByGUID

Original Method

Extended Method

Fig. 8. The extensibility of PIIM Framework.

78

I.-C. Wu, S.-H. Hsieh / Automation in Construction 23 (2012) 7186

Fig. 9. The organizational relationship of the NTU project.

Company A examined the project information and veried its feasibility in order to implement corrective or preventive actions to bring the project into compliance with the project management plan. Company A then submitted the planned project information to NTU. After NTU approved the plan, the planned project information was then baselined and functioned as a reference for controlling the project. The Project Construction Phase consisted of the processes used to complete the work as dened in the project management plan to accomplish the project's requirements. Therefore, Company A controlled the execution of the project, monitored the ongoing project activities against the project management plan and evaluated the project performance baseline in this phase. Additionally, Company A was responsible for effective communications with other project participants as information was needed to facilitate easy, effective, and timely decisions. These work items were required for achieving the project objectives on-time and within the planned scope and budget. In the Project Completion Phase, the main focus of Company A included administrative activities such as project information management, project information maintenance, and information sharing.

Company B
Construction Firm

Company A
Construction Management Firm

NTU
Owner

Summit Planned Project Information Collect Project Information

Preparation

Examine and Verify Project Information Update Actual Project Information

Summit

Feedback

Prove

Monitor and Control Engineering Project

Construction

Evaluate Performance

Summit

Feedback

Prove

Manage Project information

End

Exchange and share Project Information

Fig. 10. The activity diagram for the NTU project.

I.-C. Wu, S.-H. Hsieh / Automation in Construction 23 (2012) 7186

79

VisPMIS

Graphic User Interface Application Service DataAdapter DataManager DataViewer DataAnalyst DataExporter

requirements for assisting Company A in achieving the objectives of the NTU project. 4.3. Design and implementation of prototype system

PIIMFramework

Data Access Interface Integrated Data Model

Fig. 11. System architecture of VisPMIS.

During this phase, Company A was responsible for ling and preserving the related project information and documents. Furthermore, the reciprocal sharing of project information would allow the receiver to use the information for their own purposes, and thus supply valueadded services. This research prototyped VisPMIS according to the

The system architecture of VisPMIS is shown in Fig. 11. All of the application functions implemented by VisPMIS are based on the PIIM Framework. Core project management services and the complicated functionality of project information integration, management and visualization were implemented in the PIIM Framework. In this paper, we focused largely on the application service and the graphical user interface (GUI). With regards to the application service, we implemented the application functions as they were needed by Company B, using the services supported by the PIIM Framework. Furthermore, we designed and implemented the GUI to meet the requirements according to the specic workow and needs of Company B. Object-oriented technology was employed for the development of VisPMIS. Therefore, the conceptual model was developed using the Unied Modeling Language (UML), depicting the process and logical

frmAddScope

Return

Return
frmDocument frmNd frmCreateNdProject Return
frmAddDocument

frmSimulation

DataViewer

frm4DView

frmMonitor

frmEarnedValue

frmMain

frmQuery

DataAnalyst
Return frmControl Return Return Return

DataManager

frm3D

frmExportIfcxmlFile

DataAdapter

DataExporter

frmSchedule

frmExportGmlFile

frmCost

frmExportIfcFile

Fig. 12. The main class diagram of VisPMIS.

80

I.-C. Wu, S.-H. Hsieh / Automation in Construction 23 (2012) 7186

Fig. 13. The main graphic user interface of VisPMIS.

views of VisPMIS. As depicted in Fig. 12, VisPMIS was established on the ve easy-to-use modules of the PIIM Framework, while eighteen GUI forms were available for the user to control and manipulate VisPMIS. The implementation of VisPMIS was carried out in the

Microsoft Visual Basic .NET environment. For developing VisPMIS, we only wrote several hundred lines of code and created eighteen GUI forms. It is clear that using the PIIM Framework can reduce development efforts.

Integration
IFC
frmImport 3D

DataAdapter

Ifcxm l

Save

ProjectXML

frmImport Schedule Save Save IFC 2x3 frmImport Cost Integrated Data Model

xls

Fig. 14. Importing data examples.

I.-C. Wu, S.-H. Hsieh / Automation in Construction 23 (2012) 7186

81

4.4. Demonstration The NTU project was used as an example to test and demonstrate the functionality of VisPMIS. Fig. 13 illustrates the main graphical user interface (GUI) of VisPMIS. The upper-most menu bar (as indicated by Frame A) consists of menu items for quick access to the functions of VisPMIS. The second upper-most toolbar (as indicated by Frame B) is used to control the action of the construction simulation. The right-hand side window (as indicated by Frame C) is used to separately manage various construction project data. The left-hand side window (as indicated by Frame D) is used to facilitate the nD binding tasks. The middle window (as indicated by Frame E) is for the graphical visualization of 3D models and 4D simulation. The bottom window (as indicated by Frame F) is used to show related information for simulation and the Earned-Value method analysis. The main set of functionalities designed and implemented in VisPMIS are demonstrated in this section. The various functionalities of VisPMIS can be categorized as follows: (1) Multi-Dimensional Integrated Data Model, (2) Project Management, (3) Visualization, and (4) Information Exchange and Sharing. 4.4.1. Multi-dimensional integrated data model VisPMIS provides functions to assist the user in establishing the multi-dimensional integrated data model. Figs. 14 and 15 show that developers can build the application using the DataAdapter of PIIM Framework as a base for receiving and integrating the different data formats into the integrated data model and GUI. Subsequently, in order to bind a multi-dimensional integrated data model, the user can simply drag and drop selected 3D objects, schedule objects, cost objects, quality objects, and document objects to the target scope item in the left-hand side window, as shown in Fig. 16. The

user is then able to save the data to create a new integrated data model. 4.4.2. Project management VisPMIS provides many useful functions for project management. As shown in Fig. 17, a user can employ the Earned Value Management (EVM) method to control and monitor the project. Users can specify the date and name of the project using frmMain, and then use the DataAnalyst function of the PIIM Framework to calculate the related value for earned-value management. Finally, the outcomes will be shown in frmControl and the related graphs will be shown in frmEarnedValue as depicted in Fig. 18. As illustrated in Fig. 19, VisPMIS provides a main panel for users to monitor the project status according to the specied date. Additionally, the system can provide a multi-data-view for users to search for and obtain the necessary and relevant data in a more direct and efcient way, as shown in Fig. 20. 4.4.3. Visualization In VisPMIS, by representing the project information visually, the project manager can visualize at once the present stage of the ongoing project, and communicate and coordinate project tasks accordingly with other project participants more efciently. As shown in Fig. 21, VisPMIS can provide 1D (text), 2D (graph and chart), 3D (3D Model), 4D (3D + Time Simulation), and nD (multi-dimensional information) visualization for ease of communication among the project participants. A visualization example for construction simulation is shown in Fig. 22. After DataManager is used to collect the required data for construction simulation, the results will be shown in frmSimulation. Users can use the panel in frmMain to control the construction simulation which is shown in frm4DView.

Fig. 15. Importing a 3D model.

82

I.-C. Wu, S.-H. Hsieh / Automation in Construction 23 (2012) 7186

Fig. 16. nD model binding.

4.4.4. Information exchange and sharing Especially notable is the information exchange and sharing functionalities in VisPMIS. VisPMIS is not only able to integrate the project data, the information of the integrated data model can also be exported in three different le formats for data exchange and sharing, such that the value of the system is increased and the application of the integrated information is extended. As Fig. 23 shows,

VisPMIS can export GML data format, which is used for integration with 3D-GIS geo-data. 5. Conclusion This paper has detailed the development processes of the PIIM Framework, explaining the requirements of multi-dimensional information

Management
DataAnalyst
frm4DView frmControl Return

Return CCalEarnedValue

frmMain

DataViewer

frmEarnedValue

ZedGraph

Fig. 17. Development example for earned-value management.

I.-C. Wu, S.-H. Hsieh / Automation in Construction 23 (2012) 7186

83

Fig. 18. GUI of earned-value management.

Fig. 19. Monitoring project work.

84

I.-C. Wu, S.-H. Hsieh / Automation in Construction 23 (2012) 7186

integration, management, and visualization of engineering projects, which we found by way of the questionnaire survey and industry interview. According to these requirements, we designed and developed the PIIM Framework. Furthermore, the PIIM Framework is an application framework which is constructed on the basis of objectoriented techniques. Developers can take advantage of programming techniques to include features such as modularity, reusability and extensibility to facilitate the development and the maintenance of application systems. This research prototyped a VisPMIS system based on the PIIM Framework. VisPMIS can assist construction management rms in controlling and managing engineering projects. We compared the efciency gains from using VisPMIS for the NTU project. There are ve major benets from applying VisPMIS in the NTU project, as follows: (1). Single Window In the original method, the project data was stored on different systems. Users were required to retrieve the specic data from the specic systems and to arrange the data again to access the useful information. By establishing an integrated data model, engineering project information can be stored in an intact and complete manner in VisPMIS. Users can then nd the necessary and relevant data needed from the integrated data model through a single window. Furthermore, VisPMIS provides many query functionalities for assisting users to retrieve data quickly and accurately. (2). Communication In the original method, the teams primarily used paper-based views of project information to share information with each other in project meetings. This leads to difculties in communicating relationships between project information for project management and decisions making. In VisPMIS, by representing

the project information visually, the project manager can grasp more quickly the present status of the ongoing project, and communicate and coordinate more efciently with other project participants. (3). Multi-Data-View Much of engineering document management is le-based management, which is stored and preserved in the project les on the le server or in the le cabinet. As such, data retrieval through les is very slow and inefcient. With the support of data query functions from the Multi-Data-View in VisPMIS, users can nd necessary and relevant data in a more direct and accessible way. (4). Project management VisPMIS provide additional functionality to assist project management. Firstly, through performance evaluation and statistical analysis, project management teams are assisting in measuring the project progress accurately. Secondly, through establishing an integrated data model, this helps the project management team to retrieve useful project information more efciently. Thirdly, through the various visualization techniques, this helps the project management team to communicate with the other project participants more clearly. (5). Information Exchange and Sharing In VisPMIS, the information of the integrated data model can be exported to different le formats for data exchange and sharing, such that the value of the system is increased and the application of the integrated information is extended. VisPMIS cannot only integrate and manage engineering project information for construction management rms, it can also effectively assist in controlling and monitoring the engineering project. All of the characteristics mentioned above are inherited from the PIIM

Queried Object Related Schedule Object Related Cost Object

Related Document Object

Fig. 20. Multi-data-view.

I.-C. Wu, S.-H. Hsieh / Automation in Construction 23 (2012) 7186

85

1D & 2D Visualization

3D Visualization

4D Visualization
Fig. 21. Visualization types of VisPMIS.

nD Visualization

Framework. Through the feasibility study of VisPMIS, we found that the PIIM Framework cannot only assist construction management rms to develop their own systems to solve the issues of multi-

dimensional information integration, management, and visualization in engineering project, it can also decrease programming effort on developing a useful system from less to more.

Visualization
frmMain

DataManager

frmSimulation

CSimulationTable

DataViewer

frm4DView

AxOctagaModellerX

Fig. 22. Development example for construction simulation.

86

I.-C. Wu, S.-H. Hsieh / Automation in Construction 23 (2012) 7186

a) User Interface of IFC2GML

b) Display of IFC2GML Output (GMLModel)

Fig. 23. GML exporting function in VisPMIS.

References
[1] H. Kerzner, Project Management: A Systems Approach to Planning, Scheduling, and Controlling, Wiley, USA, 2005. [2] D.H. Zhong, J.S. Zhang, New method for calculating Path Float in Program Evaluation and Review Technique (PERT), Journal of Construction Engineering Management 129 (5) (2003) 501506. [3] W.D. Cottrell, Simplied Program Evaluation and Review Technique (PERT), Journal of Construction Engineering Management 125 (1) (1999) 1622. [4] M.B. Woolf, Faster Construction Projects with CPM Scheduling, McGraw Hill, USA, 2007. [5] WALKTHRU PC Version 1.0, 3D Simulation, User's Manual, Bechtel's Software, 1991. [6] Parsons Brinckerhoff, Corporate IT Tools SCDOT Construction and Resource Management, PB Network, 2004 Issue 58. [7] Parsons Brinckerhoff, Building Our Future, PB Network, 2005 Issue 60. [8] T. Nagasaki, K. Banno, K. Wakasugi, A. Yasaka, T. Tanimura, LINCS (Kajima Linkage of Information for New Construction System), Proc. of the Eight International Conference on: Computing in Civil and Building Engineering, Stanford, CA, United States, 2000, pp. 518525. [9] F. Pena-Mora, S. Tanaka, Information technology planning framework for Japanese General Contractors, Journal of Management in Engineering 18 (3) (2002) 138149. [10] S.H. Hsieh, C.S. Chen, Y.F. Liao, C.T. Yang, I.C. Wu, Experiences on development of a 4D PLANT construction simulation system, Proc. of the 11th International Conference on Computing in Civil and Building Engineering, Montreal, Canada, 2006. [11] S.H. Hsieh, C.S. Chen, Y.F. Liao, C.T. Yang, I.C. Wu, Construction director: 4D simulation system for plant construction, Proc. of the Tenth East Asia-Pacic Conference on Structural Engineering and Construction (EASEC-10), Bangkok, Thailand, 2006, pp. 135140. [12] I. Howell, B. Batcheler, Building Information Modeling Two Years Later Huge Potential, Some Success and Several Limitations, The Laiserin Letter, May 2005. [13] S. Azhar, W.A. Carlton, D. Olsen, I. Ahmad, Building information modelling for sustainable design and LEED rating analysis, Automation in Construction 20 (2) (2011) 217224. [14] J.P. Zhang, Z.Z. Hu, BIM- and 4D-based integrated solution of analysis and management for conicts and structural safety problems during construction: 1. Principles and methodologies, Automation in Construction 20 (2) (2011) 155166. [15] Z.Z. Hu, J.P. Zhang, BIM- and 4D-based integrated solution of analysis and management for conicts and structural safety problems during construction: 2. Automation in Construction 20 (2) (2011) 167180. [16] T. Liebich, J. Wix, IFC Technical Guide, IAI, 2000. [17] M. Halfawy, T. Froese, Component-based framework for implementing integrated architectural/engineering/construction project systems, Journal of Computing in Civil Engineering 21 (6) (2007) 441452. [18] J. Plume, J. Mitchell, Collaborative design using a shared ifc building model learning from experience, Automation in Construction 16 (1) (2007) 2836. [19] H. Schevers, J. Mitchell, P. Akhurst, D. Marchant, S. Bull, K. McDonald, R. Drogemuller, C. Linning, Towards digital facility modelling for Sydney Opera House using IFC and semantic web technology, ITcon 12 (2007) 347362.

[20] Y. Song, A. Hamilton, H. Wang, Built environment data integration Using nD modeling, ITcon 12 (2007) 429442. [21] A. Lee, M. Betts, G. Aouad, R. Cooper, S. Wu, J. Underwood, Developing a vision for an nD modelling tool, Proc. of CIB W78 Distributing Knowledge in Building Conference, Aarhus, Denmark, 2002. [22] A. Lee, S. Wu, A. Marshall-Ponting, G. Aouad, J.H.M. Tah, R. Cooper, nD modelling a driver or enabler for construction improvement? RICS Research Paper Series 5 (6) (2005) 116. [23] D.G. Treicher, Are you missing the boat in training aid? Film and AV Communication 1 (1967) 1416. [24] K.M. Liston, M. Fischer, J. Kunz, Designing and evaluating visualization techniques for construction planning, Proc. for the 8th International Conference on Computing in Civil and Building Engineering, Stanford University, CA, USA, 2000, pp. 12931300. [25] K. Liston, M. Fischer, J. Kunz, Requirements and Benets of Interactive Information Workspaces in Construction, CIFE Technical Report, Stanford University, 2001. [26] L.M. Sheppard, Virtual building for construction projects, IEEE Computer Graphics and Applications 24 (1) (2004) 612. [27] E. Collier, M. Fischer, Visual-based scheduling: 4D modeling on the San Mateo County Health Center, Proc. of the Third Congress on Computing in Civil Engineering, ASCE, Anaheim, CA, USA, 1996, pp. 800805. [28] K. McKinney, J. Kim, M. Fischer, C. Howard, Interactive 4D-CAD, Proc. of the Third Congress on Computing in Civil Engineering, Anaheim, CA, USA, 1996, pp. 383389. [29] K.M. Liston, M. Fischer, J. Kunz, 4D annotator: a visual decision support tool for construction planners, computing in civil engineering, Proc of International Computing Congress, Boston, USA, 1998, pp. 330341. [30] K.W. Chau, M. Anson, J.P. Zhang, 4D dynamic construction management and visualization software: 1. Development, Automation in Construction 14 (4) (2005) 512524. [31] K.W. Chau, M. Anson, D.D. De Saram, 4D dynamic construction management and visualization software: 2. Site trial, Automation in Construction 14 (4) (2005) 525536. [32] T. Huang, C.W. Kong, H.L. Guo, A. Baldwin, H. Li, A virtual prototyping system for simulating construction processes, Automation in Construction 16 (5) (2007) 576585. [33] N. Dawood, S. Sikka, Measuring the effectiveness of 4D planning as a valuable communication tool, Journal of Information Technology in Construction 13 (2008) 630636. [34] IFCsvr, http://tech.groups.yahoo.com/group/ifcsvr-users/ 2011. [35] Q.W. Fleming, J.M. Koppelman, Earned Value Project Management, Project Management Institute, USA, 2000. [36] D.H. Roger, Warburton, A time-dependent earned value model for software projects, International Journal of Project Management 29 (8) (2011) 10821090. [37] J. Pajares, A. Lopez-Paredes, An extension of the EVM analysis for project monitoring: the cost control index and the schedule control index, International Journal of Project Management 29 (5) (2011) 615621. [38] I.C. Wu, S.H. Hsieh, Transformation from IFC Data Model to GML data model: methodology and tool development, Journal of the Chinese Institute of Engineers 30 (6) (2007) 10851090.