Picture Exp_14 Picture Exp_15 Picture Exp_16 Picture Exp_17 Picture Exp_18 Picture Exp_19 Picture Exp_20

Improving fire safety of buildings by simulation in the cloud

Executive Summary

Fire safety of buildings is of relevance for each European citizen. Who does not want to feel save when shopping in a mall or visiting a theatre? In case of an emergency event, people have to get out of the building in a fast, streamlined but not chaotic manner, sprinklers have to be placed in the planning and construction phase so to keep fire under control or even extinct it, fire fighters have to have sufficient access routes and water supply, etc. This all should be simulated and optimized before starting the building process.

The market size in Spain (Europe) for fire simulation tools and services is estimated to amount for € 1.5 million (€ 20 million).

The goal of this experiment is to improve fire safety designs in the building sector using CFD. Specifically this means to integrate a CFD tool called CYPE-FDS into the CloudFlow platform for detailed fire simulation scenarios focused on the building design industry. As a show case, a real shopping centre located in Spain will be used for a complete fire safety design process. The expected technical impact is an improvement of the complete workflow of the fire safety design stage, a reduction in time for the model preparation and results analysis, and higher accuracy in the prediction compared to the traditional process through leveraging the open source solver FDS.

Economically this will show the following effects: a) an increased number of fire-safety design projects, b) reduced costs of the fire-safety facilities due to optimization (30 percent cost reduction as an average, approximately € 80.000 in a case similar to the one solved in this experiment), and c) reduced costs for hardware and software for users applying fire simulation due to a per-per-use concept for HPC resources and open source software.

Currently, fire protection analysis and design is based on scalar and prescriptive models (simple and fast approach) according to codes like CTE DB SI 6 code, Eurocode (EN 1992-1-2:2004 and EN 1993-1-2:2005). But these models do not supply detailed information about the movement of smoke and the temperature evolution. Although the usage of CFD tools would be possible, the technology is not commonly used in industry because of the hardware and software costs, the limited number of CFD specialists and the time constraints for defining the fire safety design. In the reference case, this represents approximately 30 percent of the total time for the complete building design (structural and facilities). The complete fire safety design using CFD tools must be solved in no more than three weeks in order to be competitive with prescriptive models.

The challenges lie in providing a user-friendly tool with fast response times using HPC resources to facilitate fire simulations for AEC engineers that are not experts in fire simulation software. The approach is to cloudify the open source solver FDS to obtain the necessary accuracy for the fire protection analysis and design. CYPE-FDS, the pre-post tool, which facilitates the case definition, will be available in the workflow through a virtual machine. Several fire scenarios could be executed in parallel thanks to HPC resources inside CloudFlow. Finally, the results obtained in the case experiment will be summarized in a Best Practice Guideline for FDS use in building industry. This guide could be used to establish strategies for estimating a good balance between the cell size, the solving time, and the accuracy of results; HPC efficiency (number of cores, solving time reduction, simulation cost ratios); and numerical considerations in order to define subdomains.


CYPE, the ISV partner, will be able to offer new cloudified solutions for their software to new and existing clients. Furthermore, improvements in the CYPE FDS workflow are realized due to the use of HPC resources. Adopting the same procedure for other software, e.g. structure analysis, energy efficiency and construction management, CYPE will be able to increase the cloudbased product offerings allowing not only for solving more complex problems in the domain of fire simulation but also other domains.

Cottés and Itecam, both end-users in the experiment, profit from the user interface improvements, especially for pre- and post-processing the model. The training time for the CFD tool can be reduced significantly by providing application-specific tailored functionality. Additionally, access to HPC resources is streamlined by the workflow and does not require special know-how on the user side.

Concerning the fire safety facilities design the end users benefits from the increasing number of fire scenarios that could be simulated. Finding the optimum design is speed up by 30-40 percent due to the usage of the cloud solution because it is accessible anytime, has no idle time and runs with good stability and low risk in simulation interruption.


CYPE will address a bigger target market with the expectation of 2,000 clients within one year after the experiment completion. Furthermore it is likely to start 200 projects with existing clients, one year after the experiment completion, with a perspective of 400 projects in the third year. Concerning new clients, CYPE expects 100 projects in the first year and 300 projects in the third year, respectively. The financial benefit with a total income for the first year is estimated by € 57,000 and by € 192,000 in the third year in software license sales. CYPE is planning to employ two new software developers.

At the moment only 20 percent of the Fire Safety Design Projects from the end user Cottés require the usage of CFD tools. This corresponds to € 350,000 - € 400,000 per year given a total revenue of € 3.5 -4.0 Mio. Using the cloudified solution of the experiment, Cottés expects to double the number or projects resulting in a budget of approx. € 750,000 per year. Cottés is also expecting to reduce the costs of the active and passive mechanisms and gaining more probability of winning the tendering process. Today, Cottés trades only in the Spanish market, developing 20 fire safety design projects in 2016. Gaining competitiveness through the cloudified version, the number of projects could be doubled. Furthermore, Cottés customers would benefit from obtaining optimized solutions in shortened time period, with a 30-40 percent cost reduction, as an average € 80,000 per project.

Another benefit is the potential access to new markets like major civil engineering construction projects (airports, tunnels) and the forensic engineering sector. Another sector could be opened if insurance companies will ask for a CFD solution in order to know the origin of a fire. If 0.1 percent of these fire scenarios required a detailed analysis through CFD simulations, Cottés would increase its number of projects in a range from 15 to 25. After the CloudFlow Project, Cottés is planning to employ two more CFD specialists in fire simulation and one engineer expert in Fire Safety Codes.

In 2016 ITECAM has developed four CFD projects for private companies using commercial software such as ANSYS-CFX, Solidworks Flow, etc. focusing on CHT analysis, pressure drop simulations, and coupled multi-physics problems. Thanks to the “Fire in the Cloud” application experiment, ITECAM expects to expand its CFD simulation capabilities to other areas, such as: HVAC analysis, FSI simulation under fire scenarios, human evacuation analysis, etc. At the end of the third year, ITECAM is planning to consolidate a specific CFD department with 2-3 engineers, dedicating 50 percent of its activity to FDS simulations. In 2017, ITECAM expects to achieve revenues close to € 50,000 through the use of “Fire in the Cloud”, collaborating with Cottés and CYPE in consulting activities, benchmarking, and training some of our associated companies. In 2020 the CFD department is expected to achieve revenues close to € 250,000, approximately 45 percent of this amount would be generated through FDS activities.

Arctur being an infrastructure provider expects to have mostly an economic impact form this experiment. This is reflected through the increased sale of Cloud and HPC resources. The increase of the sales follows the increase of new users using the services at offer.

Fire in The Cloud