[Music] health safety and environment management otherwise known as hse is extremely important to industrial operations and especially in oil and gas traditionally in hse management data is mined from historical incident reports recorded in standard forms with check boxes and pull-down menus these records are manually analyzed to generate new regulations advisories best practices and training this is extremely time consuming for safety supervisors and often leads to incomplete results where only known risks are identified further complicating things semantic searches or keyword classification only identify partial events and oftentimes return inaccurate responses due to the inability to take context into account for example not knowing if feet is meant as a body part or as a measurement with vast amounts of data coming in safety supervisors have trouble keeping operations as safe as possible leaving the company and its people vulnerable there is so much information which is captured in the near misses and the accident reports that you have in your organization and we have multiple ways to improve the safety records of your organization using artificial intelligence sparkcognition's deep nlp product uses natural language processing to automatically analyze structured and unstructured data recognizing meaningful patterns in the data by analyzing incident reports in text format the deep nlp product can automatically identify new risks enabling supervisors to apply corrective action before an accident occurs we can reclassify the entire database to mine for new risk on a daily basis in one case spark cognition worked with a large national oil and gas operator that received thousands of hazard observation cards per month from different sites the customer wanted to automate the analysis of these observation cards and extract actionable insights to help improve workplace health and safety spark cognition used the deep nlp product to ingest the data from the observation cards extract each hazard type assign the likelihood of a hazardous event occurring and provide a summary of potential risk areas the deep nlp product also provided a platform for supervisors to create assign and close action items related to each hazard the results were surprising the customer discovered that driving into animals was a significant risk to their employees which was previously unknown the data revealed that workers were driving into all kinds of animals what ai has been able to do is to put all those animal in the cluster and say you know what we have a lot of animals we have a lot of driving when you put it together driving into animal is a problem thanks to the deep nlp product the customer saw a reduction in the number of accidents involving animals in a recent study the overall lost time injury rate was 0.24 per million hours worked with participating companies reporting roughly 22 000 days of work lost through injuries the deep nlp product for hse helps organizations identify their top hazardous activities locations in compliance and deviation from hse standards when you understand the risk you can fix it the risk which is going to injure your people or break some equipment as a risk you do not know about with artificial intelligence we can identify those new risks very early in the process this knowledge enables the organization to focus their hse activities to minimize the probability of incident and lost time injury to learn more about how the deep nlp product can improve hse management for your operations contact us at info sparkcognition.com you