Semantic data models for built environment
Lead Research Organisation:
University College London
Department Name: Bartlett Sch of Env, Energy & Resources
Abstract
The PhD studentship will commence with a gap analysis and documentation of barriers on technical and non-technical aspects that include: comparison and evaluation of expressiveness of open vocabularies for data publishing, publishing data and attribution using different license models, technologies supporting data and privacy sharing and protection, all with a view to the built environment applications. Following the initial gap analysis, the work will select appropriate use cases and research on ways to removing technological barriers. Of particular interest is the development of data ingestion methodologies, data completeness characterisation, and quality checking tools. This is very good opportunity for talented students to work with a network of academics in the built environment field giving unparalleled access to data, knowledge and professional contacts.
Parallel to the development of the world-wide web the vision and associated technologies for the publishing of data have evolved and the web of data has emerged for sharing structured data sets. Even though fundamental technological components are available like, for example, the Resource Description Framework (RDF) and Web Ontology Language (OWL), there is still an evolving discussion on relevant vocabularies and the ability to easily export, publish and make discoverable such data sets; technological barriers include issues related to: knowledge representation, data fusion, integration and standardization, data quality and validation. In addition to technological barriers, there are issues related to data ownership, attribution and usage that currently are not adequately addressed.
For built environment applications, a number of open vocabularies are emerging for data sharing: these include the ifcOWL ontology based on the ISO 16739:2013 standard, the SAREF ontology, SimModelOWL and others. The possibility of use of such information in linked-data contexts, where information from multiple domains can be linked using semantic linking tools, allowing for more effective data exploration and data ingestion can have a transformative potential for built environment research. Data requirements exported by national regulations (e.g. EU Public procurement directive 2014/24) will increase the availability of data in open formats.
Parallel to the development of the world-wide web the vision and associated technologies for the publishing of data have evolved and the web of data has emerged for sharing structured data sets. Even though fundamental technological components are available like, for example, the Resource Description Framework (RDF) and Web Ontology Language (OWL), there is still an evolving discussion on relevant vocabularies and the ability to easily export, publish and make discoverable such data sets; technological barriers include issues related to: knowledge representation, data fusion, integration and standardization, data quality and validation. In addition to technological barriers, there are issues related to data ownership, attribution and usage that currently are not adequately addressed.
For built environment applications, a number of open vocabularies are emerging for data sharing: these include the ifcOWL ontology based on the ISO 16739:2013 standard, the SAREF ontology, SimModelOWL and others. The possibility of use of such information in linked-data contexts, where information from multiple domains can be linked using semantic linking tools, allowing for more effective data exploration and data ingestion can have a transformative potential for built environment research. Data requirements exported by national regulations (e.g. EU Public procurement directive 2014/24) will increase the availability of data in open formats.
Organisations
Studentship Projects
| Project Reference | Relationship | Related To | Start | End | Student Name |
|---|---|---|---|---|---|
| EP/N509577/1 | 01/10/2016 | 24/03/2022 | |||
| 1828831 | Studentship | EP/N509577/1 | 06/02/2017 | 05/02/2021 | Jackline Kibe |
| Description | An increasing demand in professional connectivity consequently fuelled by the pressures of rapidly evolving technological advances is placing great demands on industries to deliver more interactive data and knowledge management systems that enable the extraction of new knowledge autonomously and facilitate decision making processes. This is more so apparent in the Facilities Management (FM) industries where stakeholders are faced with the challenge of improving and standardising the quality of the information/data they have at their disposal in order to meet the day-to-day operational needs. The underlying premise is that current asset information (AI) / data management processes have remained largely unstructured and siloed leading to incredibly time-consuming and inefficient day-to-day maintenance practices. The challenge lies in the fact that over an asset's lifecycle, the inevitable introduction of new variables such as new compliance requirements, changes in ownership, environmental & sustainability requirements and economic conditions mean that there are continuous changes to AI requirements and data. This research project hypothesises that use of ontologies can be used to provide a centralised global context to the information stored in the Asset Information Model (AIM) to effectively improve FM internal asset maintenance internal auditing processes. The purpose of this study so far has been to expand knowledge and understanding about the key factors that undermine effective management and sharing of asset information (AI) in FM. The key objectives of the study are: to investigate the concepts of semantic & linked data technologies, to investigate the concepts of FM asset management data and associated data exchange formats, to examine the constructs of information/data exchange in FM data environments, to examine the principles product/asset data and metadata and identify the data sets critical to the FM operation and maintenance and to identify the key asset data/information exchange challenges in an FM centric environment via case study investigation. Key issues identified in from published literature and the case study can be summarised to include; (i) un-reliable asset data in the CAFM systems, (ii) the loss of historical maintenance information i.e. remedial actions, drawings, (iii) the loss of day-to-day maintenance information i.e. task sheets, service sheets, (iv) the lack of maintenance schedules for various assets, (v) outdated maintenance information i.e. drawings, asset registers, (vi) sub-standard maintenance information i.e. unsigned/undated documents, (vii) non-compliant maintenance tasks i.e. outdated instruction sets and (viii) inadequate asset information. Further investigations are focussed on developing a framework in which a seamless and dynamic link between FM disciplines where data can be shared to suit the users' requirements. This link would allow FM's access all the critical asset information regardless of whether it is stored in siloed & disparate systems such as financial systems, CAFMs, BIMs, EMS etc. or whether it is in a web based database. The ontology paradigm has emerged as light weight approach to integrate data from heterogeneous data sources by generating links between data sets. The proposed ontology will be used to identify the critical use IA asset information as determined from an asset maintenance FM perspective. |
| Exploitation Route | While the Architectural, Engineering and Construction industries are slowly beginning to embrace and emerging design breakthroughs and beginning to incorporate IoT ideas and technologies into project life cycle in a bid to improve efficiency of the processes, measurement of performance and transparency of the overall project, progress has been very slow in the Facilities Management industries. Although this can be attributed to many cultural and economic factors, a distinct lack of research into how semantic technologies can be used to enhance FM management processes and what the expected benefits are has also been a key deterrent to developers and investors across the globe. More funding into research, particularly in relation to addressing the problem of interoperability between the various heterogeneous applications and systems used to support facilities management would attract more interest from the IoT developers investing in integrated FM environments. |
| Sectors | Digital/Communication/Information Technologies (including Software),Healthcare,Other |
| Description | To understand the rationale to integrating semantic technologies into Facilities Management (FM) asset information management processes a case study investigation into a live FM environment was undertaken. The scope of the study was focused a PFI-funded hospital building in which the award holder is part of the team that provides facilities management services. The key objectives of the case study was to determine the scope of the asset management system by identifying the; the relevant stakeholders, the scope of their involvement to that respect, its interaction with other FM management systems and identify the associated challenges of the system specifically within a specific building. Results from the investigation highlighted a significant problem within the asset information management systems. Investigations included looking into the asset information model which serves as a for all the information generated, collected or stored to support the management of the assets owned or managed by the FM provider for the hospital. Methods used to collect structured and unstructured Hard FM data/information include; manual data collection from over 50 Sub-contractors & 12 In-house technicians, automated data collection from the BMS and smart gadgets etc. and semi-automated data collection which refers to the combination of both manual and automated data collection. According to senior FM management issues pertaining to asset information management had led to increased maintenance costs due to either over or under maintaining assets which at times resulted into downtime of hospital services and therefore hefty fines for FMs and loss or mismanagement of O&M information. The case study investigation identified three key issues; i.e. missing information (maintenance service records), outdated information (drawings, log sheets etc.) and substandard information (inaccurate data, un-signed documentation etc.). Interviewees unanimously attributed the issue to the lack of a centralised asset information management system. The results from the case study investigation were informally used as part of the organisations initiative to improve FM asset management within the building. Significant efforts are currently being undertaken by the management team to move the FM operation processes from paper based systems into "paperless maintenance" program using mobile devices. |
| First Year Of Impact | 2019 |
| Sector | Digital/Communication/Information Technologies (including Software),Healthcare,Other |
| Impact Types | Economic |