Using AI and Computer Vision to Improve Workplace Safety & Comply with OSHA’s COVID-19 Guidelines

Much has been written about China implementing nearly draconian measures to control the spread of Covid-19 which helped to quickly restart manufacturing segments of their economy.  The reality is not quite that simple. The New York Times article of May 13th “Work Essentials in China: Masks, Soap, and Vigilance” makes it clear that there will be no return to pre-COVID normal in China or much of the world.1  Chinese workers are told to wash their hands both before and after handing documents. However, even in authoritarian China, there is a lack of clear, central guidance as to what the new rules should look like.  But some of the rules are obvious – maintain social distancing, wear face masks, avoid touching paper documents, or entering data on keypads.  This is where AI and computer vision can help to improve workplace safety and can follow new WHO and OSHA guidelines.

The World Health Organization (WHO) published guidelines on March 3rd warning employers that they needed to take action even if the virus had not yet arrived or if infection rates had dropped off. “Make sure your workplaces are clean and hygienic, Surfaces (e.g. desks and tables) and objects (e.g. telephones, keyboards) need to be wiped with disinfectant regularly. Why? Because contamination on surfaces touched by employees and customers is one of the main ways that COVID-19 spreads.” 2

OSHA has published its first guidance on preparing the workplace for a life in the age of COVID-19.3 For the first time it classifies occupations by their risk of infection.

Lower exposure risk jobs are those that do not require contact with people known to be, or suspected of being, infected with SARS-CoV-2 nor frequent close contact with (i.e., within 6 feet of) the general public. Workers in this category have minimal occupational contact with the public and other coworkers.

Medium exposure risk jobs include those that require frequent and/or close contact with (i.e., within 6 feet of) other people who may be infected with SARS-CoV-2 even if they show no symptoms.

High exposure risk jobs are those with high potential for exposure to known or suspected sources of COVID-19. This includes Healthcare delivery and support staff and medical transport workers.

Very high exposure risk jobs are those with high potential for exposure to known or suspected sources of COVID-19 during specific medical, postmortem, or laboratory procedures that involve aerosol generation or specimen collection/ handling. This includes healthcare workers, and laboratory personal exposed to COVID-19 patients.

What is most concerning in these classifications is that the only truly low exposure risk is for those lucky enough to work alone or at home. Everyone else is at risk given that COVID-19 can easily shed or spread the virus for at least two weeks before an individual shows any visible symptoms, including running an elevated temperature.  There is also a wide medical consensus that the virus will not go away for least a few years. The history of the 1918 pandemic has taught us that the second wave will be more deadly than the first.

SOCIAL DISTANCING

Computer vision is a great tool to monitor social distancing compliance in workplace environments while protecting personal privacy. The examples below show varying degrees of social distancing in the workplace, from safe and compliant to severe violations of policy. In the two cases on the right, it is clear that distances are too short to prevent community/people transmission. As a default, Atollogy obfuscates and obscures individual identities, but this can be configured depending on privacy policies.

 

PAPER HANDLING

Processes as simple and seemingly harmless as handing paperwork from one person to another or keying in data on tablets and terminals will now be seen as potential vectors to spread the highly contagious and deadly virus. The less drivers and equipment operators are required to key in data or fill out forms, the greater their personal safety. There are indicators that COVID-19 and related viruses can live up to 72 hours on surfaces, so the fear of physical contact is bound to continue even after infection rates subside.1 The Chinese suggestion to wash hands before and after handling any document is not realistic or sustainable. The better solution is to eliminate the paperwork.

 

Computer vision can not only avoid virus transmission by paper but improve efficiency of work centers. In one customer example, operators enter work order data on personal tablets. Computer vision is used to validate the start and end times for each job and thus calculate efficiency. This can also be used to count quantities produced.

CONCLUSION

It is reasonable to assume that we are not likely to ever return to a pre-COVID world in the workplace  With a virus that is ten times as deadly as the flu and has pushed most all economies into a major recession and possibly a depression, strong countermeasures will be needed for the foreseeable future.  While neither WHO nor OSHA guidelines are yet legal regulations, any workplace that ignores them will invite a litigious nightmare – think the meatpacking industry.  Regardless of regulations and fear of litigation, workers will seek out employers who follow best practices in workplace safety. AI-based computer vision is a proven and cost-effective technology providing 24X7 monitoring and alerts to workplace safety which will help to reduce violations and improve employee satisfaction.

References

  1. https://www.nytimes.com/2020/05/12/business/coronavirus-china-work-safety.html
  2. https://www.who.int/docs/default-source/coronaviruse/getting-workplace-ready-for-covid-19.pdf
  3. https://www.osha.gov/Publications/OSHA3990.pdf

 

Author:

 

Anthony Tarantino, PhD

Adjunct Professor, Santa Clara University – Lean Six Sigma and Supply Chain

Six Sigma Master Black Belt, Certified Scrum Master, CPIM (APICS), CPM (ISM)

Senior Advisor to Atollogy

Tony@atollogy.com

atarantino@scu.edu

562-818-3275