This chapter discusses the system requirements in which the tasks to be performed by the rule-based system have been clearly specified. Furthermore, the chapter introduces Semantic Web tools to be used in developing the system. It is important to note that construction labour cost ontology is implemented on its two main components. Firstly, the knowledge model is implemented in an ontology editor where editing of the knowledge model is conducted. Secondly, the knowledge model is implemented in a rule-based editor where SWRL rules are edited to facilitate reasoning over the construction labour cost model. Key in this chapter the transformation of the UML knowledge model of Chapter 8 to OWL. The process of converting UML to OWL has been discussed in Abanda (2011). The effort converting UML to OWL will not be duplicated here. The chapter presents an illustrative example of how a rule is implemented in the main development tool, Protégé-OWL. some screenshots of key aspects including top ontology concepts, an excerpt of the OWL ontology and some sample rules and queries have been presented.
System Requirements
The establishment of system requirements is imperative in the development of any knowledge-based systems as this specifies the tasks to be performed by the knowledge base. The task to be performed by the knowledge base depends on the knowledge requirements specified by the CommonKADS methodology. The purpose of this section is to discuss the various strategies used to build a rule-based application for the management of construction labour cost information. In order to develop an application it was imperative to implement the labour cost ontology in a software environment. To decide on which software to use, it was imperative to establish the requirements of the ontology knowledge-based system. How and what will the labour cost ontology be used for? Therefore the following requirements have been established:
- a.
Provision of an ontology knowledge-based system where labour cost information can be stored and retrieved;
- b.
The system should support reasoning. An ontology that will support reasoning. That entails the creation of some constraints and rules that may contextualize the information related to the role of class and properties for reasoning;
- c.
Provision of the possibility of the labour cost ontology to adapt and evolve with minimal disruption. New ontologies can be defined and added incrementally without the need for the redesign of the environment.
To meet the requirements (a-c), the following software and components tools have been used:
- a.
Protégé-OWL 3.4.4: This is an ontology development editor developed by Stanford University, USA. It is an open-source tool that enhances end-users skills in creating, visualizing, and updating ontologies. It is very extensible and can accommodate other plug-ins that can be used in developing other applications. The plug-ins used in support of protégé-OWL are:
- b.
SWRLTab: This is a protégé plug-in and editor that facilitates the writing of SWRL rules;
- c.
JessTab: This is a protégé plug-in that allows the use of Jess (a rule language) and protégé together;
- d.
Pellet 1.5.2: Pellet is an OWL 2 reasoner which provides standard and cutting-edge reasoning services for OWL ontologies. Although pellet currently exists in different versions, pellet 1.5.2 is the version that has been incorporated in protégé-OWL 3.4.4, the ontology editor chosen for this study.
The implementation of the labour cost ontology based on the methodology examined in section 4 in the protégé-OWL 3.4.4 environment is presented in Figure 1. The figure depicts the main labour cost ontological components including some classes‟ instances, properties and SWRL rules.
Figure 1.
Labour Cost ontology in protégé-OWL 3.4.4 screenshot
The main ontology components captured by protégé-OWL and shown in Figure 1 are the classes, instances, properties and rules. In section 7, scenarios are presented depicting how the labour cost ontology is queried using SWRL rules highlighted in Figure 1.