The world has been dealing with the COVID-19 pandemic for several months now. While some countries have managed to flatten the curve, others are still struggling to control transmission. Regardless of the state of the curve, many nations are faced with tough decisions related to safely reopening the economy to save their countries from financial depression. While normalcy seems like a distant possibility, many essential workers have come to terms with going back to work and risk getting infected.
Like other companies, ESI has mandated precautions and strives to implement safety procedures intended to control and prevent the virus from spreading. While these measures are a good starting point, our team in India felt the need to evaluate how effective they really would be at ensuring the safety of their workplace and their colleagues.
The Coronavirus crisis, which we have all been experiencing for several months, is distressing a lot of habits and has resulted in new issues to be addressed by companies. Employee safety has always been key, but when the danger is invisible and circulating in the air, the topic appears to be more complicated. Just like other industrial and technological players, we seek to provide as many people as possible with solutions to help rebuild, secure, and reassure. Simulation is an ideal tool that enables testing a large range of scenarios without endangering anyone – which is exactly what we did to reassure our very own employees. Our expertise allowed us to create realistic scenarios of droplet behavior in enclosed and ventilated spaces. We are pleased to provide our input and to continue being faithful to our values: being as close as possible to our customers to help them achieve their goals, including the most challenging ones, and to always work in the best interest of our own people as ‘One ESI’.Anshul GuptaCOO, ESI India Business Operations
Modern offices are designed to ensure that they protect people from external elements like pollution and dust to maintain an atmosphere that is comfortable for its occupants. Unfortunately, this also means less direct ventilation. If fresh air is introduced, it can be done only through the central air conditioning system, which is always mixed with return air for energy efficiency. A configuration like this is good for both the occupants and the climate as it reduces the energy consumed by the air conditioning systems. However, these modern workspace designs fail to consider and address risks such as the spread of a virus, as such scenarios were not the primary concern.
According to the World Health Organization (WHO), COVID-19 is an airborne disease. Naturally, this raises concerns over the safety of workspaces and triggers the following questions:
The list goes on.
Our engineers got to work on this right away, digging into how simulation can address these concerns for others and for their own colleagues. They replicated various real-life scenarios and conducted virtual tests to locate the areas in the given workspaces that are relatively safe to occupy; this means learning about the air movement pattern and how it can be controlled.
They considered the following scenarios:
To represent a typical office or factory scenario, we considered individual office spaces, a common work area, a kitchen/dining area, and an assembly line.
Naturally, personal offices are high on the list of problematic workspaces. Some might think of their office as a haven, closing the door and hoping visitors stay at bay. On the other hand, a small, closed-off area could raise fears of lingering germs lying dormant in the stagnant airspace. So, how safe are we really and how safe are the people visiting our space?
To make the simulation setup realistic, we used one of our own office layouts, replicating the HVAC ducts exactly. We considered flow rates in line with actual flow rates.
We based this simulation on the cough of a person walking into the office (the person standing). The cough was modeled considering the density of water vapor, the theoretical distribution of cough particles, the number of cough particles, as well as their velocity based on actual test data . The natural air movement from the open window in the ventilation model (far right, Fig. 2) was modeled with a wind velocity of 0.5 m/s, consistent with a light breeze.
In the ventilation model, most of the particles drifted away from the cabin 20 seconds after the cough because of the natural ventilation (open window & door).
The simulation with air-conditioning (center image, Fig. 2) also yielded encouraging results: the return duct succeeded in attracting particles towards it, as particles are very light averaging around 1 microgram. An even closer look reveals that most of the light particles are sucked into the return duct, which suggests that adding a filter to catch them after they enter the duct would be a probable solution (since most central air-conditioning units run on partial recirculation mode). Filters can be changed often, however, there will be additional pressure drops across these filters, so the power consumption may increase slightly.
Finally, the simulation with no air conditioning, meaning little movement or stagnant air, proves to be the worst-case scenario. The particles float around the room even after several minutes.
Open workspaces or cubicles are typically the most populated areas in the office. They also tend to be large with little to no ventilation opportunities. Due to the current social distancing norms, we modeled this scenario with only one person in each cubicle.
Fig. 3 shows us that it is very difficult to draw inferences from this scenario. As the particles are small (only a few microns) and light, they are easily carried by air currents and make their way any and everywhere. Except for the pockets where air stagnates, most of the floating particles will eventually make it to a return duct. Therefore, the consensus would be to avoid these areas of stagnating air and consider new seating arrangements.
The ISO surface shown in Fig. 4 (the envelope of air the same age as initial air in the environment) indicates the area where there is the least amount of fresh air; in other words, the areas where the age of the air is relatively higher. These areas should also be avoided.
The kitchen/dining area might be the most feared space in today’s current climate as people can’t depend on masks to protect them while enjoying a meal.
As explained in the open work area, the iso-surface in Fig. 6 clearly shows the stagnant pockets of air that should be avoided while seated.
After running several simulations to reassure our employees in India about how to safely return to the office, we come to, perhaps, the most essential area for all manufacturing companies or OEMs – the assembly line. How could the industry safely send their teams back to work?
The assembly line is where most goods are produced, some of which are essential to our survival. Assembly lines are usually an enclosed, large space where air movement is bound to be non-uniform due to obstructions and partitions, etc. Also, these spaces are typically dynamic environments, with people moving around, machines running, and parts & components being passed around, which makes social distancing extremely challenging.
For this simulation, we mimicked a standard automotive assembly line where there is a lot of heavy lifting and movement throughout the day. This also typically means heavy breathing, which can make a person more susceptible to either transmit or receive the virus than the average person, and while performing especially physically demanding jobs, wearing a mask may not be advisable.
The simulation considered a scenario where two people are working on a car’s chassis. The ambient conditions contained powerful drafts of air coming from the blowers. Under these conditions, we simulated one person coughing to assess the spread of particles.
Even in this simulation, the ISO surface in Fig. 11 shows the amount of stagnant air, which should be avoided at all costs. Air velocity vectors around the chassis are also weak as can be seen in Fig. 10, meaning that the cough particles are going to take a long time to drift away with the air current. However, these conditions can be improved by either changing the operating conditions of the blowers or by using additional blowers or localized fans closer to the workers to ensure that there are no stagnant air pockets around them and that fresh air is being circulated.
From the various scenarios that we simulated, we observed that Virtual Prototyping gives great insight into the problem and helps us find a customized solution. With the help of these tests, you can ensure proper ventilation of your office space, and most importantly, validate the safety of your employees.
For our team in India, we are happy to report that this simulation experiment validates the local guidelines that were mandated and have armed local management with the right information needed to ensure the safety of the employees, which is always our top priority. As the COVID-19 situation in India hasn’t improved yet, working from home remains the default for our India team. However, management is now more confident than ever about the measures or the steps that should be taken to ensure the safety of our employees as a result of the detailed simulation results.
We are proud to share that our team was awarded the COVID-19 prize of the 2020 Simulation and Artificial Intelligence Award by “L’Usine Digitale” for the work they did on this project.
Watch this segment on BMF TV (in French only) where they discuss how ESI's solutions help to visualize COVID-19 in open spaces.
Local Experts team: Venkat Ramana Eaga (CFD), Ravi Kumar Ajjampudi (CFD), Bharath Isandra Govindappa (Virtual Reality), Mr. Sandeep Patil (CFD, Pre-Processing)
Quantity and Size Distribution of Cough-Generated Aerosol Particles Produced by Influenza Patients During and After Illness. Journal of occupational and environmental hygiene. 9. 443-9. 10.1080/15459624.2012.684582. Lindsley, William & Pearce, Terri & Hudnall, Judith & Davis, Kristina & Davis, Stephen & Fisher, Melanie & Khakoo, Rashida & Palmer, Jan & Clark, Karen & Celik, Ismail & Coffey, Christopher & Blachere, Francoise & Beezhold, Don. (2012).