Digital Architecture for Humanitarian Design in Post-Disaster Reconstruction

Architecture Dated:  March 23, 2015
submit to reddit
0 Comment(s)
Source: huftonandcrow.com

Digital tools and computational design processes are rapidly changing Architecture. Nonetheless their applications in humanitarian design remain under researched. Generative algorithmic design is particularly useful in humanitarian design and post disaster reconstruction. Firstly, the extreme conditions in these contexts pose many constraints that can be parametricised and form the basis of a parametric design. Secondly, optimal use of scarce resources is enabled by integrating these interrelated performance requirements. Thirdly, a robust model definition afforded through parametric modelling enables a mass customised design to adjust for different site and user requirements, and most importantly it allows improvements in subsequent design based on community evaluation. Over the past decade computational architectural methods such as BIM (building information modelling), parametric modeling and GA (genetic algorithm) have been increasingly employed to generate design. Nonetheless their applications have not yet filtered through to applications in the field of humanitarian design. This article describes the context and a proposal for applying Architectural computing in humanitarian design in remote areas of developing countries that are the most technologically challenged. A system of digital tools, in particular parametric modelling and BIM, can be devised to optimize the exemplar design for site and project specific needs, and reduce time and cost required in the overall design. Starting from the premise that the constraint-bound context of humanitarian design particularly suit the rule-based nature of parametric systems, and that any level of automation to reduce time required for architects in the field would lead to optimal use of project funds.

Theoretical Issues Of Generative Design And Humanitarian Design:

Parametric Design:
Digital tools such as generative modelling are increasingly used in architectural design, but their major applications lie in a few high investments buildings or unbuilt proposals due to the complex geometric forms enabled by the software. Many question if such form driven design process is ethical (Ostwald 2009). Barrow and Kumar (2007) argued that obsession with ‘free-form’ generation has undermined the powerful function of advanced modelling to perfect performance and mass-customisation -these functions, rigorously exercised in industrial design, should be applied to housing design. That is to retain the principles of achieving higher buildability from simplification, standardization, and modularisation, but moving from mass production to mass customization with the new digital tools available.

Humanitarian Design:
The critique on the approach to social housing described above overlaps with some of the challenges and opportunities presented to humanitarian design – they are both limited by resources, and often are given a one-size fit-all solution multiplied over a large area but insensitive to site or user. Nonetheless the high number of units per prototype offers the economy of scale to develop a system that adapts for specificities unique to each unit. Humanitarian design in a low-tech context involves many complex issues outside of typical construction. In the case of post-disaster emergency relief, speed and safety is the priority. This is where standardised, prefabricated solution is often seen as the most appropriate. However prefabrication also involves importing high-tech or expensive components that are foreign to developing countries. These conflicts with the other essential criteria for a successful design – one that local people can replicate and maintain with the technologies available to them. Though justifiable for the case of immediate response to emergency relief, the potential of customization through parametric design has not been fully explored for transitional or permanent applications. Currently the most successful model in post disaster reconstruction focuses on community involvement to achieve best value recovery outcomes (Lawther 2009).The underlying theorem is that the more the recovery relies upon local resource, the quicker the community will be able to move to self-sustainability, and thus from recovery to normalcy. Sullivan(2003) identifies the link between involvement of the community in post disaster recovery and mental recovery, noting that it ‘alters their status from passive pawns in the process, to once again active and contributing directors of their own destiny’. By extension, the success of a building design is impinged on parameters that encompass those psychological and social needs. Post-disaster reconstruction and development work in the third world posses some of the most extreme scenarios and challenging constraints. Firstly there is the limitation of resources in terms of funding, material, time, foreign expertise and accessibility of the sites. These interrelated constraints can be translated into the parameters for generative design to allow optimum use of those scarce resources. Further, a robust model definition afforded through parametric modelling enables a mass customised design to adjust for different site and user requirements, and allow a design that gives a greater sense of ownership to those recovering from trauma. Secondly few organizations in development work have well-established protocols to revisit their works after completion to test if the design works according to intent. Even if post occupancy evaluation was carried out, this is often not translated into subsequent set of design due to the lack of an integrated platform to retain this knowledge or the tools to quickly regenerate new design in a feasible way. Parametric systems offer a unique tool in support of communities and local people to manage these constraints. At present such constraints are usually managed by foreign aid workers at high cost. Although there is often a capacity building component in foreign aid assistance this usually becomes a case of a local person with some technical expertise executing a limited array of tasks within the overall delivery. Parametric modeling could be employed at this stage to limit the costs. In consultation with communities, local technical participants, government and foreign aid workers could set up a parametric process that responds to this local consultation and provides an ongoing feedback loop. Local technical participants could then be tasked to work within this process and charged with varying various parameters within the model, thus giving greater local ownership over the process. An additional benefit would be the continued refinement of a process as opposed to a new process each time a project starts.

Current Gap in Research:
We should aim to address the issues above and open up a field for further research. The objective is to optimize, customize, build, evaluate, improve and record. We should start with a small project and a targeted set of tools and expand on the complexity of the building and the array of tools over time. Typically, assessment of site and needs, building design and supervision of construction are led by foreign experts with a high cost base, until sufficient capacity can be built up locally. Any means to alleviate reliance on foreign experts would allow funds to be spent more directly on construction and ensure programs continue after disengagement with international agencies. Engagement terms of field experts are often limited to a year. Moreover the pressing nature of field work inhibits proper documentation and transfer of the knowledge gained from field experience to local staff members and succeeding field workers. Currently the most successful delivery model involves international aid agencies designing and building exemplar prototypes as a way of transferring knowledge to communities. Funding and drawings are supplied to communities for the next build. However the prototype often requires further modifications to suit site and project specific requirements. The field architects and engineers must then manually repeat the calculations and documentation, lengthening the process.
In summary, a successful design would require economical use of simple locally obtained material that can be built easily and quickly with the most basic tools and skills, and allow participation by everyone in the community. The system should be designed to adapt various site and user requirements, and to absorb unforeseeable circumstances such as shortages in certain materials, to ensure long term viability of the projects, designs should also be easily replicated by communities with minimal external assistance.

Parametric opportunities in humanitarian design:
The constraints and user requirements discussed above often call for innovative solutions that have not been tested. However post disaster reconstruction and development typically replicates hundreds of buildings from one prototype in many different sites over time - this raises opportunities to evaluate the design and construction process and incorporate users’ feedback to improve subsequent sets of designs, not dissimilar to the rigour practiced in industrial design. Similarly, application of digital technologies such as parametric modeling lacks built examples. The few built works tend to be expensive one-offs and require a long time to design and construct, limiting application of lessons learnt to new projects. In contrast, development work from one prototype has the potential to allow these technologies to be tested and refined on comparable basis over time. It would be much more meaningful, if research and development in architectural computing are geared towards real projects with real impact, than temporal installations and objects.

Technique:
The values from the requirement calculation spreadsheet are fed into the design parameter spreadsheet which stores all the parametric values for the digital model in Rhinoceros®. This spreadsheet is bi-directionally linked with Grasshopper® - allowing the model to be updated by values in the spreadsheet or directly controlled with sliders in Grasshopper. The link enables instant updating of the quantity and cost of material and fuel, and estimates of how many deliveries are required to transport all the material, as shown in the BOQ. Study of the model and adjustment to the parameters give the user the ability to attain a final desired outcome through informed decisions. Various options can be captured for rapid prototyping, allowing further study and comparison in reality. The model and the BOQ also inform the user of potential areas of cost savings, as well as opportunities for increasing amenity with a very slight increase in cost. Another important aspect is recording of the changes in scripted model and the iterative variations. The Grasshopper script is labeled with explanations on the rationale behind decisions made for the design model. VIP DIGITAL ARCHIPUNCTURE - Flowchart summarising the system and Changes to the parametric values resulting from post occupancy evaluation were also tagged on the script, to allow an integrated platform for
 

Discussions:
The Design Definition – Adaptability and Optimization:
The system enables a high level of automation throughout various stages of the project. Time spent performing tasks such as requirement calculations, BOQ and documentation are significantly reduced, and errors that could occur during the otherwise manual procedures are minimized. The direct link between design parameters and cost allows instant and accurate cost implications to be analysed. Potential savings can be identified and a balance struck between such parameters as material quantities, structural analyses, transportation and desired amenities, without compromising the design, and the volumetric calculations prevents over provisioning of material.  As a result of the visual variable control afforded through the Grasshopper interface, mixed with the information-rich input from a spreadsheet, the designer is able to quickly determine an optimum solution for a given scenario, work within the computed constraints and develop a sensitive design solution. More factors can be considered in this way than if done without the aid of the developed digital tools. Iterative studies of building proportions and formal relationships of the building elements, produces a more refined architecture. This process would have been prohibitively time consuming if performed manually, but crucial for a sense of unique ownership for the beneficiaries. Better visualisation tools and rapid prototyping also enable the designer to make more informed decisions for the final design.

Innovation in Construction Documentation:
Parametric modelling facilitates innovation in construction documentation to better suit the scale of the building and the local context. The “graphical 3D step by step instructions” is easily understandable, regardless of prior training or literacy level. The laser cut model further assists in visualising how individual elements are assembled. These visual aids assist in understanding and encourage community participation. Ease of construction provided by the “one to one scale templates” for slabs and junctions allow local carpenters and people from the community to construct a more complex building than was previously feasible.

Retention and Accumulation of Knowledge:
The spreadsheets outlining the rationale for establishing user requirements, parametric design variables, the BOQ, and any additional data such as post occupational evaluations are all interlinked with the parametric model. When the evaluation suggests a change to a parameter value, or a change to the script, it is labeled within the parametric script.
This acts as one platform for the retention and accumulation of knowledge for even temporary field volunteers, and allows disparate disciplines such as structural engineering and sanitation to operate on the same model. Further, this coherent body of growing knowledge allows management staff from non-construction related background to make more informed decisions. Preliminary feedback on the system from field workers indicates this feedback loop was valued the most highly, and should be further refined for the next stage of the research - as success does not result from the production of a predetermined product, but a system designed for refinement of appropriate outcomes.

Conclusion:
Best practice humanitarian design differs fundamentally from the provision of design within the context of the developed world. Despite both require a high degree of professionalism however the focus should not be on a marketable product but on a process that empowers recipients, such as local communities, technical participants and local governance structures. Without local engagement and buy in into the development process, such programs are rarely able to build enough momentum to enact broad ranging improvements. This article started with the aim of investigating how the application of ‘high tech’ digital technology could support such processes in the ‘low tech’ constraint-bound context of humanitarian architecture. In the course of bridging the gap in generative modelling and humanitarian design we found ourselves in an emerging field of research, and perhaps addressed some of the criticisms raised on the current use of digital tools by opening up the field to work in more pragmatic but challenging contexts. Our developed methodology emphasizes address of contextual constraints, user requirements, and learning from experience. Building upon existing best practice developed from fieldwork, a system of digital tools was introduced in various stages of the project where opportunities arose. The system improves accuracy and shortens the time required to perform calculative tasks. The parametric definition allows application of the prototype to adapt a range of possible scenarios. Iterative studies linked to cost analysis enabled optimising solutions otherwise not feasible within time constraints. Construction documentation tailored for local needs improves the procurement process and will encourage community participation. Further, the model allows a growing, coherent body of knowledge for permanent and temporary aid staff to ensure highly skilled outcomes. Through the use of generative design, it is possible to build a system that facilitates the ongoing process with the main objective of empowering local people and responsive to evolving requirements. This is achieved by firstly recording the results of local consultation which then supports a feedback loop, and then entrusting the implementation of the monitored design at a local level. At anytime foreign expertise or external evaluation can be provided, however most importantly the system ensures that this input contributes as an evolution to the system. With lessons learnt and local requirements always addressed, the product becomes the embodiment of what local people require. And who could possibly understand their own development requirements better than those who strive to develop.

Future Research:
Effective means of further optimisation in areas such as material and structural stability sunlight and ventilation, should be implemented. More importantly, based on the preliminary comments from field experts, we will focus on how to better enhance the feedback loop in streamlining the evaluation assessments and the iteration of subsequent designs, and the recording of how the design evolves. For example, site selection based on wind and terrain data supplied locally in addition to GIS information; more detailed assessment criteria. Evaluation of the effectiveness of the system onsite should be carried out in the next step of this research.
This article is loosely based on the paper presented by Wendy K.Yeung and Jeremy Harkins
Y Architecture, University of New South Wales wendy@wendyyeung.com, jeremyh@unsw.edu.au 

References:
1. Sinclair, C. and Stohr, K.,Architecture for Humanity Staff, Design Like You Give a Damn:Architectural Responses to Humanitarian Crisis, Metropolis Books, New York, 2006.
2. Barrow, L., Kumar, S. and Arayedh, S.: Emerging Technology – Dilemma and Opportunities in Housing, Proceedings of the 12th International Conference on Computer Aided Architectural Design Research in Asia, CAADRIA 2007, Nanjing, 301-308.
3. Caldas, L.G. and Norford, L.K., Shape Generation Using Pareto Genetic Algorithms: Integrating Conflicting Design Objectives in Low-Energy Architecture, International Journal of Architectural Computing, 2003, 1(4).
4. Chakraborty, S.,Automated Generation of Residential Roomlayout within a Constrained Covered Area, Proceedings of the 8th International Conference on Computer Aided Architectural Design Research in Asia, CAADRIA 2003, Bangkok, 2003, 85-100.
5. Hecker, D., Dry-In House:A Mass Customized Affordable House for New Orleans, Proceedings of the 10th Iberoamerican Congress of Digital Graphics, SIGraDi 2006, Santiago de Chile, 2006, 359-362.
6. Jinuntuya, P. and Theppipit, J.,Temporary Housing Design and Planning Software for Disaster Relief Decision Support System, Proceedings of the 12th International Conference on Computer Aided Architectural Design Research in Asia, CAADRIA 2007, Nanjing, 2007, 639-644.
7. 30 Wendy K.Yeung and Jeremy Harkins 02_Yeung  08/06/11  8:39 am  Page 307. Kaga,A. and Miyagawa,A., Construction of a Participatory Community Space Design System, 25th eCAADe Conference Proceedings; Frankfurt am Main, 2007, 99-106
8. Lawther, Peter M., Community involvement in post disaster reconstruction – case study of the British Red Cross Maldives Recovery Program, International Journal of Strategic Property Management, 13(2), 2009, 153-169.
9. Love,T., Between Mission Statement and Parametric Model,Available from:The Design Observer Group.
10. Ostwald, M.J., Evaluating Digital Architecture: Ethics and the Auto-Generative Design Process, Ethics and the Built Environment 2009 Conference Proceedings, Adelaide, 2009.
11. Pena-Mora, F. and Mills, J.W., Component-based architecture for online collaborative disaster relief mission planning environments, CIDAC, 3(2), 2001.
12. Sener, S.M. and Torus, B., Container Post Disaster Shelters – C-PoDS, 27th eCAADe Conference Proceedings;The New Realm of Architectural Design, eCAADe, Istanbul, 2009, 599-604.
13. Sullivan, M., Integrated recovery management:A new way of looking at a delicate process, The Australian Journal of Emergency Management, 18(2), 2003, pp. 4–27.
14. Wen, K. and Chen,W.: 2004,Application of Genetic Algorithms to Establish Flooding Evacuation Path Model in Metropolitan Area, Proceedings of the 9th International Conference on Computer Aided Architectural Design Research in Asia, CAADRIA 2004, Seoul, 2004, 557-570.

Sinclair, C. and Stohr, K., Architecture for Humanity Staff, Design Like You Give a Damn: Architectural Responses to Humanitarian Crisis, Metropolis Books, New York, 2006.
Barrow, L., Kumar, S. and Arayedh, S.: Emerging Technology - Dilemma and Opportunities in Housing, Proceedings of the 12th International Conference on Computer Aided Architectural Design Research in Asia, CAADRIA 2007, Nanjing, 301-308.
Caldas, L.G. and Norford, L.K., Shape Generation Using Pareto Genetic Algorithms: Integrating Conflicting Design Objectives in Low-Energy Architecture, International Journal of Architectural Computing, 2003, 1(4).
Chakraborty, S., Automated Generation of Residential Roomlayout within a Constrained Covered Area, Proceedings of the 8th International Conference on Computer Aided Architectural Design Research in Asia, CAADRIA 2003, Bangkok, 2003, 85-100.
Hecker, D., Dry-In House: A Mass Customized Affordable House for New Orleans, Proceedings of the 10th Iberoamerican Congress of Digital Graphics, SIGraDi 2006, Santiago de Chile, 2006, 359-362.
Jinuntuya, P. and Theppipit, J., Temporary Housing Design and Planning Software for Disaster Relief Decision Support System, Proceedings of the 12th International Conference on Computer Aided Architectural Design Research in Asia, CAADRIA 2007, Nanjing, 2007, 639-644.
Kaga, A. and Miyagawa, A., Construction of a Participatory Community Space Design System, 25th eCAADe Conference Proceedings; Frankfurt am Main, 2007, 99-106.
Lawther, Peter M., Community involvement in post disaster reconstruction - case study of the British Red Cross Maldives Recovery Program, International Journal of Strategic Property Management, 13(2), 2009, 153-169.

Love, T., Between Mission Statement and Parametric Model, Available from: The Design Observer Group.
Ostwald, M.J., Evaluating Digital Architecture: Ethics and the Auto-Generative Design Process, Ethics and the Built Environment 2009 Conference Proceedings, Adelaide, 2009.
Pena-Mora, F. and Mills, J.W., Component-based architecture for online collaborative disaster relief mission planning environments, CIDAC, 3(2), 2001.
Sener, S.M. and Torus, B., Container Post Disaster Shelters - C-PoDS, 27th eCAADe Conference Proceedings; The New Realm of Architectural Design, eCAADe, Istanbul, 2009, 599-604.
Sullivan, M., Integrated recovery management: A new way of looking at a delicate process, The Australian Journal of Emergency Management, 18(2), 2003, pp. 4-27.
Wen, K. and Chen, W.: 2004, Application of Genetic Algorithms to Establish Flooding Evacuation Path Model in Metropolitan Area, Proceedings of the 9th International Conference on Computer Aided Architectural Design Research in Asia, CAADRIA 2004, Seoul, 2004, 557-570.

Liked this article? Click here to consult Shashank Ninawe for your project or design dilemma.

submit to reddit