Deconstruction and Recovery Information Modelling (DRIM): A Tool for identifying and reclaiming valuable materials at end-of-life of Buildings
Lead Research Organisation:
University of the West of England
Department Name: Faculty of Business and Law
Abstract
More than 50,000 buildings are demolished yearly in the UK leading to huge demolition waste that ends in landfill (Power,
2014). It is noted that demolition waste comprises significant proportion of valuable building materials that could be re-used
for new constructions or refurbishment if recovered properly. However, no such tool currently exists that can help in
identification of valuable building materials for reuse & recycling. The overall aim of this project is to develop an
intelligence-based tool called Deconstruction and Recovery Information Modelling (DRIM) that will enable identification of
reusable and recoverable building materials at end-of-life of a building. DRIM Tool will enable: (i) production of
deconstruction plan; (ii) simulation of deconstruction process; (iii) production of deconstruction protocols during demolition
of the building to enable efficient recovery; (iv) improved demolition waste collection schemes. The tool is aimed at both
new and existing buildings sector. The Tool will use innovative technologies that include ontologies, NoSQL and big data
analytics to capture and predict end-of-life properties and value of building materials. It is about circular economy in the
construction industry.
Methodology and Plan: This is a 24 Months (M) project of 8 Work Packages (WP): (i) WP1 - Data collection on Materials
properties using workshops with Designers and Suppliers - Month 1 - 6 (6M); (ii) WP2 - Creation of Semantic Store with
End-of-Life Material Properties - Month 4 - 9 (6M); (iii) WP3 - Creation of Big Data Analytics Simulation Platform - Month 7 -
12 (6M); (iv) WP4 -Prototype DRIM Development - Month 9 - 12 (4M); (v) WP5 - Full DRIM Development - Month 13 - 21
(9M); (vi) WP6 - Security & Testing of DRIM tool - Month 22 - 23 (2M); (vii) WP7 - Exploitation and Dissemination - Month 4
- 24 (21M); and (viii) WP8 - Project Management - Month 1 - 24 (24M). - (Please see Appendix B for Gantt Chart).
Key Milestones, Deliverables & Realistic Timelines: WP1 - Gigantic Dataset of Material Properties @ M6; WP2 - Database
storage of End-of-Life Material Properties @ M9; WP3 - Big Data Analytics based Simulation Platform @ M12; WP4 -
DRIM Prototype @ M12; WP5 - DRIM Full System @ M21; WP6 - Packaged DRIM Tool @M23; WP7 - Exploitable &
Dissemination Outputs @ M4 to M24; and WP8 - Quarterly Project Reports & Meetings.
Clear Management Reporting Lines: Lara Ayris from Waste Plan Solutions Ltd. (WPS) will project manage and coordinate
project resources, with the support of Project Administrator and Exploitation Manager for wider roll-out. Project
Management Board (PMB) will be set to drive project, monitor project progress and provide relevant advice. Using
PRINCE2 methodology, Task managers for each WP will (i) manage day-to-day project activities (ii) meet with their WP
team on weekly basis (iii) meet monthly (iv) report quarterly to the PMB about work progress . - (Please see Appendix B for
Project Management Structure).
Rival Solutions: None of the existing waste tools within the industry (i.e. ArchiCAD, Revit, SMARTWaste, WRAP Netwaste,
etc.) has deconstruction and material recovery functionality. The DRIM tool is therefore unique within the industry. It will
provide a simulation platform to benchmark the whole-life sustainability of designs in terms end-of-life re-usable, recyclable
and recovered materials.
Alternative R&D Strategies: Concurrent Engineering (CE) model which is based on parellelization of tasks (Work
Packages) will be used as a R&D approach for this project, as compared to traditional waterfall model where tasks are
carried out sequentially. CE will therefore enable R&D completion within 2 years of project duration.
2014). It is noted that demolition waste comprises significant proportion of valuable building materials that could be re-used
for new constructions or refurbishment if recovered properly. However, no such tool currently exists that can help in
identification of valuable building materials for reuse & recycling. The overall aim of this project is to develop an
intelligence-based tool called Deconstruction and Recovery Information Modelling (DRIM) that will enable identification of
reusable and recoverable building materials at end-of-life of a building. DRIM Tool will enable: (i) production of
deconstruction plan; (ii) simulation of deconstruction process; (iii) production of deconstruction protocols during demolition
of the building to enable efficient recovery; (iv) improved demolition waste collection schemes. The tool is aimed at both
new and existing buildings sector. The Tool will use innovative technologies that include ontologies, NoSQL and big data
analytics to capture and predict end-of-life properties and value of building materials. It is about circular economy in the
construction industry.
Methodology and Plan: This is a 24 Months (M) project of 8 Work Packages (WP): (i) WP1 - Data collection on Materials
properties using workshops with Designers and Suppliers - Month 1 - 6 (6M); (ii) WP2 - Creation of Semantic Store with
End-of-Life Material Properties - Month 4 - 9 (6M); (iii) WP3 - Creation of Big Data Analytics Simulation Platform - Month 7 -
12 (6M); (iv) WP4 -Prototype DRIM Development - Month 9 - 12 (4M); (v) WP5 - Full DRIM Development - Month 13 - 21
(9M); (vi) WP6 - Security & Testing of DRIM tool - Month 22 - 23 (2M); (vii) WP7 - Exploitation and Dissemination - Month 4
- 24 (21M); and (viii) WP8 - Project Management - Month 1 - 24 (24M). - (Please see Appendix B for Gantt Chart).
Key Milestones, Deliverables & Realistic Timelines: WP1 - Gigantic Dataset of Material Properties @ M6; WP2 - Database
storage of End-of-Life Material Properties @ M9; WP3 - Big Data Analytics based Simulation Platform @ M12; WP4 -
DRIM Prototype @ M12; WP5 - DRIM Full System @ M21; WP6 - Packaged DRIM Tool @M23; WP7 - Exploitable &
Dissemination Outputs @ M4 to M24; and WP8 - Quarterly Project Reports & Meetings.
Clear Management Reporting Lines: Lara Ayris from Waste Plan Solutions Ltd. (WPS) will project manage and coordinate
project resources, with the support of Project Administrator and Exploitation Manager for wider roll-out. Project
Management Board (PMB) will be set to drive project, monitor project progress and provide relevant advice. Using
PRINCE2 methodology, Task managers for each WP will (i) manage day-to-day project activities (ii) meet with their WP
team on weekly basis (iii) meet monthly (iv) report quarterly to the PMB about work progress . - (Please see Appendix B for
Project Management Structure).
Rival Solutions: None of the existing waste tools within the industry (i.e. ArchiCAD, Revit, SMARTWaste, WRAP Netwaste,
etc.) has deconstruction and material recovery functionality. The DRIM tool is therefore unique within the industry. It will
provide a simulation platform to benchmark the whole-life sustainability of designs in terms end-of-life re-usable, recyclable
and recovered materials.
Alternative R&D Strategies: Concurrent Engineering (CE) model which is based on parellelization of tasks (Work
Packages) will be used as a R&D approach for this project, as compared to traditional waterfall model where tasks are
carried out sequentially. CE will therefore enable R&D completion within 2 years of project duration.
Planned Impact
The targeted Audience include: Contractors and their supply chains, public and private clients, sub contractors, architects,
engineers and the academic community. Our strategies to engage the target audiences are as follows:
A. Dissemination:
1. Demonstration Projects: Five demonstration projects to serve as case studies in workshops. This would be video
recorded and uncovered at a project dissemination workshop at the end of the project.
2. Public Website: A public website would be made available at www.DRIM-Tool.co.uk, to be used as the main vehicle of
dissemination and interaction with the public who seeks information about the project and its areas of work.
3. Press Release: A press release, prepared by the consortium, would be launched by the kick-off date of the project and at
the end of the project announcing the project outputs and its benefits.
4. Leaflet: A public leaflet describing the DRIM Tool project would be prepared, which will be used for the presentation of
the project in main events (conferences, workshops, fairs, and so on), at both within the UK and Europe.
5. Newsletter: Quarterly Newsletter presenting the project's progress to the "outside world", containing sections that
Editorial, Under the Spotlight, Inside DRIM Tool, Looking Outside, and What's Next.
6. Participation in Target Events: The project consortium would be targeting major events within the construction industry
particularly the ones organised by RIBA, CIOB, RICS, ICE, ISructE, CIBSE, Constructing Excellence, UKCG, NFB, CPA,
among others. Some of these organisations are already member of the consortium. Their support would also be enlisted to
distribute the project leaflet to their member organisations.
7. Social Media Channels: To some extent, the Social Media channels to be used will be subject to a small-step trial-and
evaluation approach. It is not fully decided which channels existing today are the most suitable ones for the DRIM Tool
Project, and furthermore, new channels will evolve as the project runs. However, a few channels are identified as
(potential) starting points, which are listed below.
(i) Blog: A blog would be launched, hosted at the main DRIM Tool website. The blog is positioned to be a more informal
channel, with posts mainly written by individuals in the project. The posts are normally positioned as the viewpoint of the
individual, and not as the viewpoint of the project or of a project partner, when the author belongs to DRIM Tool Project.
(ii) Conversational Social Media Channels: A number of Social Media channels that are more suitable for a conversational
context are being identified, which include LinkedIn groups, Google+, etc.
(iii) Viral Marketing: The DRIM Tool Project will strive to create screencasts related to as many published papers and other
presentations as possible. These screencasts (video format), will be stored and published on established "web 2.0"
services, which include YouTube, Slide share and various photo sharing sites.
8. Academic Beneficiaries: Specifically, the project aims to engage academic community through the following activities:
Conference Papers, Research Papers, Special Interests Group, Guest Editing Special Issues, and Edited Monographs.
The academic Principal Investigator will lead these initiatives.
engineers and the academic community. Our strategies to engage the target audiences are as follows:
A. Dissemination:
1. Demonstration Projects: Five demonstration projects to serve as case studies in workshops. This would be video
recorded and uncovered at a project dissemination workshop at the end of the project.
2. Public Website: A public website would be made available at www.DRIM-Tool.co.uk, to be used as the main vehicle of
dissemination and interaction with the public who seeks information about the project and its areas of work.
3. Press Release: A press release, prepared by the consortium, would be launched by the kick-off date of the project and at
the end of the project announcing the project outputs and its benefits.
4. Leaflet: A public leaflet describing the DRIM Tool project would be prepared, which will be used for the presentation of
the project in main events (conferences, workshops, fairs, and so on), at both within the UK and Europe.
5. Newsletter: Quarterly Newsletter presenting the project's progress to the "outside world", containing sections that
Editorial, Under the Spotlight, Inside DRIM Tool, Looking Outside, and What's Next.
6. Participation in Target Events: The project consortium would be targeting major events within the construction industry
particularly the ones organised by RIBA, CIOB, RICS, ICE, ISructE, CIBSE, Constructing Excellence, UKCG, NFB, CPA,
among others. Some of these organisations are already member of the consortium. Their support would also be enlisted to
distribute the project leaflet to their member organisations.
7. Social Media Channels: To some extent, the Social Media channels to be used will be subject to a small-step trial-and
evaluation approach. It is not fully decided which channels existing today are the most suitable ones for the DRIM Tool
Project, and furthermore, new channels will evolve as the project runs. However, a few channels are identified as
(potential) starting points, which are listed below.
(i) Blog: A blog would be launched, hosted at the main DRIM Tool website. The blog is positioned to be a more informal
channel, with posts mainly written by individuals in the project. The posts are normally positioned as the viewpoint of the
individual, and not as the viewpoint of the project or of a project partner, when the author belongs to DRIM Tool Project.
(ii) Conversational Social Media Channels: A number of Social Media channels that are more suitable for a conversational
context are being identified, which include LinkedIn groups, Google+, etc.
(iii) Viral Marketing: The DRIM Tool Project will strive to create screencasts related to as many published papers and other
presentations as possible. These screencasts (video format), will be stored and published on established "web 2.0"
services, which include YouTube, Slide share and various photo sharing sites.
8. Academic Beneficiaries: Specifically, the project aims to engage academic community through the following activities:
Conference Papers, Research Papers, Special Interests Group, Guest Editing Special Issues, and Edited Monographs.
The academic Principal Investigator will lead these initiatives.
Publications
Ajayi S. O.
(2018)
Waste-efficient materials procurement for construction projects: A structural equation modelling of critical success factors
in Waste Management
Akanbi L
(2019)
Disassembly and deconstruction analytics system (D-DAS) for construction in a circular economy
in Journal of Cleaner Production
Akanbi L
(2020)
Deep learning model for Demolition Waste Prediction in a circular economy
in Journal of Cleaner Production
Akanbi L
(2019)
Reusability analytics tool for end-of-life assessment of building materials in a circular economy
in World Journal of Science, Technology and Sustainable Development
Akanbi L
(2018)
Salvaging building materials in a circular economy: A BIM-based whole-life performance estimator
in Resources, Conservation and Recycling
Akinade O
(2017)
BIM-based deconstruction tool: Towards essential functionalities
in International Journal of Sustainable Built Environment
Akinade O
(2018)
Designing out construction waste using BIM technology: Stakeholders' expectations for industry deployment
in Journal of Cleaner Production
Akinade O
(2019)
Integrating construction supply chains within a circular economy: An ANFIS-based waste analytics system (A-WAS)
in Journal of Cleaner Production
Akinade OO
(2017)
Design for Deconstruction (DfD): Critical success factors for diverting end-of-life waste from landfills.
in Waste management (New York, N.Y.)
Akinosho T
(2020)
Deep learning in the construction industry: A review of present status and future innovations
in Journal of Building Engineering
| Description | Key findings and development are: 1) Essential functionalities of BIM-based deconstruction tools: Evidence shows that future direction for effective Design for Deconstruction (DfD) is the adoption of BIM-based approach for design coordination. After a review of extant literature on existing DfD practices and tools, it became evident that none of the tools is BIM compliant and that BIM implementation has been ignored for end-of-life activities. To understand how BIM could be employed for DfD and to identify essential functionalities for a BIM-based deconstruction tool, Focus Group Interviews (FGIs) were conducted with professionals who have utilised BIM on their projects. The interview transcripts of the FGIs were analysed using descriptive interpretive analysis to identify common themes based on the experiences of the participants. The themes highlight functionalities of BIM in driving effective DfD process, which include 'improved collaboration among stakeholders', 'visualisation of deconstruction process', 'identification of recoverable materials', 'deconstruction plan development', 'performance analysis and simulation of end-of-life alternatives', 'improved building lifecycle management', and 'interoperability with existing BIM software'. 2) Development of a BIM-based predictive model for estimating the salvage performance of buildings: A BIM-based predictive model was developed to estimate the salvage performance of buildings. The model development was based on factors that influence recoverability and reliability distribution of building materials. These factors, which were obtained from the literature were classified into two categories, i.e., those that influence reusability and those that influence recyclability. The predictive model was evaluated using three design specifications of a real-life building case study. The results show that building design with steel structure, demountable connections and prefabricated assemblies generate recoverable materials that are mostly reusable (i.e. 93% reusable, 7% recyclable). Whereas, building design with concrete structure generates recoverable materials that are mostly recyclable. The design with timber structure generates recoverable materials that are largely reusable (i.e. 65% reusable, 35% recyclable). 3) Development of deep-learning models for the prediction of end-of-life performance: Deep-Neural Network (DNN) models were developed for the prediction of demolition arisings (wastes). The input of the DNN-models are: (i) Gross Floor Area (GFA), (ii) volume, (iii) number of levels, (iv) building usage type and (i) building frame type. The overall average performance of the DNN models is 90%. The DNN-models could predict the end-of-life arisings by component types. From the distribution of the datasets used for the model development, the performance of the DNN models on residential building with masonry construction type is well above 90%. This is because the dataset comprises of 70% residential building information and 79% masonry building. 4) Development of a BIM-based web application for deconstruction and recovery information modelling: The web application is a BIM-interfacing software tool to simulate the end-of-life of a building. Key features of the web application include: (i) predictive analytics of end-of-life scenario using DNN-models, (ii) visualisation and simulation of end-of-life and deconstruction of buildings, (iii) reporting of end-of-life arisings in terms of materials that could be reused and recycled, (iv) Autodesk-Revit integration through an Addin to support designers during design-for-deconstruction. 5) To support the goal of circular economy agenda of the UK government, i.e. keeping materials values in the economy for as long as possible, a reusability analytics tool for assessing end-of-life status of building materials was developed. The reliability analysis principle and materials properties were used to develop the predictive mathematical model for assessing building materials performance. The results of analytics show that the quality of the building materials varies with the building component. For example, in concrete buildings, the foundation component has the highest quality followed by first floor. The concrete obtainable from stairs has the least quality. As a contribution to the concept of circular economy in the built environment, the mathematical model developed provides a foundation for estimating the quality of obtainable building materials at the end-of-life based on the life expectancy of the building materials. Knowing the quality of materials that will be obtained from a demolition process will enable adequate planning for materials reuse and recycle. 6) Deep learning model for demolition waste prediction in a circular economy: The deep learning models developed in this work take advantage of the advancement in big data and machine learning algorithms to establish a meaningful relationship between the independent variables, (i.e. building architypes, building usage, building's gross floor area, number of floors and volume) and corresponding dependent variables (recyclable materials, reusable materials and landfill materials). The three DNN models developed have an average accuracy of 97% when R-squared is used as the performance metric. The independent variables contribute to the three DNN models' performance differently, for the recyclability DNN model, building usage type has the highest level of importance i.e.1.00 followed by gross floor area (0.95), number of floor (0.91), architype (0.76) and lastly volume (0.62). In the reusability DNN model, the number of floors in a building is a major determinant with 1.00 level of importance, the least important variable is the volume with 0.64 level of importance. In the landfill DNN model the number of floors in a building has the highest level of importance (1.00), followed by the gross floor area with 0.87 level of importance. The least variable that contributes to the model is the volume with 0.23 level of importance. The results from the evaluation of the DNN models with different scenarios of the case study building confirm the conclusion of previous studies (Akanbi et al., 2019, 2018; Akinade et al., 2015) that timber and steel based buildings produce end-of-life materials that are mostly reusable through either direct reuse or recycling. The present study also shows the end-of-life performance of the masonry-based building. The masonry-based building generates end-of-life arising that is 56.98% recyclable, 39.44% reusable and 3.58% waste materials to landfill. The masonry-based building performs more than three times better than concrete-based building in terms of materials reusability. The reason for the masonry performance is not farfetched, bricks and blocks which are the basic components of a masonry building come in standard sizes, which makes them readily useable in other projects without any modification. While the current results are in line with results from previous studies, the present study establishes the possibility of using basic information about building to predict the end-of-life arisings. Previous works depend on the availability of a well-defined federated BIM model of the building to estimate the end-of-life arisings. The prediction functionality provided by the DNN models in this study is particularly useful in the situation where there are little or no information about a building that is meant to be deconstructed/demolished. This is the situation with the most buildings that are due to be demolished in the UK. |
| Exploitation Route | The essential functionalities for BIM based deconstruction tools are key to improving the performances and collaborative capabilities of DfD tools. This is especially required as the industry is far shifting towards a fully collaborative digital workflow and the building deconstruction industry can benefit from this. The results also bring to the fore that visualisation capability of BIM is key to simulate and visualise building deconstruction process during the design stage. This will enable early detection of possible site operational or management issues, such as transportation logistics, waste management, scaffolding requirements, health and safety considerations. The essential functionalities will provide the needed technological support for developing the similar tools. The predictive models will be integrated into a BIM-based tool to estimate the salvage performance of buildings right from the design stage. The BIM-based tool will provide an assessment method for evaluating buildings' potential for compliance with the circular economy goals and reducing waste disposal to landfills. The web application developed as part of this project is the first of its kind in the UK construction industry. As such, it provides a roadmap to others who are interested in developing similar tools. |
| Sectors | Construction,Environment |
| Description | The DRIM-tool developed for estimating amount in tons and categories of materials is being trailing internally by an SME that undertakes Pre-Demolition audits for client. Feedbacks are being received and used to fine-tuned the solution. The tool has also been demonstrated to Costain-Skanska Joint Venture on HS2. The possibility of using the DRIM-tool for demolition phase of the HS2 project. The Costain-Skanska JV got to know about the DRIM-tool from one of the newsletter released by the project team. |
| Title | A BIM-based web application for the DRIM software |
| Description | Deep learning models for predicting the total salvage (i.e. recyclable and reusable) and disposable values of major building materials were developed. Then, a BIM-based web application was developed for the DRIM Tool to provide building end-of-life analytics and model simulation. The deep learning models were integrated into the web application for end-of-life prediction and pre-demolition report generation. The application is currently hosted on Amazon Web Service. |
| Type Of Technology | Webtool/Application |
| Year Produced | 2018 |
| Impact | The web application is a collaboration platform that allows several stakeholders to work together on a deconstruction project. Pictorial presentation of the salvage values of building materials and waste route (reusable, recyclable and landfills) was also integrated with the DRIM web platform. |
| Description | Software Demonstration at Costain-Skanska JV |
| Form Of Engagement Activity | A formal working group, expert panel or dialogue |
| Part Of Official Scheme? | No |
| Geographic Reach | Local |
| Primary Audience | Professional Practitioners |
| Results and Impact | Demonstration of the DRIM tool to three practitioners from Costain-Skanska JV work on the demolition phase of the HS2 project and to discuss trialling the DRIM tool on upcoming demolition projects. |
| Year(s) Of Engagement Activity | 2018 |