Missing Link… OSHA Combustible Dust Rulemaking Process
05 July 2011
Throughout the global manufacturing sector, the loss of life due to combustible dust incidents continues to haunt stakeholders. Headlines reappear with statistics and graphic accounts such as the recent May 20 Foxconn dust explosion in Chengdu, China.
Then in another time zone, a series of combustible dust flash fires at the Gallatin, Tennessee Hoeganaes plant in the US.
It is important to note that fatalities, injuries, and property damage are sustained from both combustible dust fires and explosions. Not just dust explosions. So now a search for the missing link….
Subsequently, a level of concern arises in the US with the current OSHA Combustible Dust rulemaking process. This is especially so where a multitude of combustible dust related fires in the manufacturing sector is ignored in the vein of normalisation of deviation. So why are fire statistics not included in the comment period and national dialogue during the development of a regulation similar to ATEX and DSEAR?
Lack of Combustible Dust Fire Statistics
Overall, the biggest problem is that the fire statistics are not being collected in a comprehensive manner. So what incident data is being utilised in the rulemaking process? It all began with a series of catastrophic dust explosion that occurred in 2003. Following these explosions, the US Chemical Safety Board (CSB) completed a 2006 Dust Hazard Study. It was through the urgent recommendations in the CSB report that OSHA initiated the combustible dust rulemaking process in 2009.
The CSB report is an excellent cause and effect analysis of combustible dust explosions and provides very informative guidance in preventing and mitigating future combustible dust incidents. Where the CSB Dust Hazard Study falls short is the paucity of incident data solely identifying 281 combustible dust incidents from 1980-2005.
In contrast, through several years of intensive research the Combustible Dust Policy Institute discovered through reviewing media accounts there are ten times more incidents occurring in the USA compared to CSB data. Most importantly, 80% of incidents are combustible dust related fires such as in dust collectors, ductwork, etc. The majority of these incidents are “near misses” with no fatalities, injuries, or property damage. In all fairness CSB is an accident investigative organization and not a research entity.
Equipment Involved in Ignition (EII)
An in-depth view in the lack of understanding the probability of occurrence and severity of consequence points to the voluntary fire incident reporting system in the US. This is referred to as the National Fire Incident Reporting System (NFIRS 5.0) administered by the US Fire Administration and funded under the Department of Homeland Security and Federal Emergency Management Administration (FEMA).
More detailed insight can be found in a 2009 fire statistics report by the NFPA Fire Analysis and Research division that utilized NFIRS 5.0 data in the 2003-2006 timeframe, "Structure Fires in Industrial and Manufacturing Properties." The results of the report noted 23 percent, or $107 million, of direct property damage occurred at industrial and manufacturing facilities where the equipment involved in ignition (EII) was unclassified. This includes all fires whether combustible dust related or not.
So what is classified? Well that would be manufacturing process equipment that provided the primary ignition source such as industrial furnaces, pumps, bearings, belts, grinders, etc. A problem arises when conveyors, mills, ductwork, dust collectors, mixers, blenders, bucket elevators, etc. do not have data elements in the Fire Modules of NFIRS 5.0 to identify if they provided the ignition source, thus they are unclassified.
Fires, Precursor Warnings to Catastrophic Dust Explosions
The most difficult aspect for facility owners and safety managers to understand is how the prevalence of non-consequential combustible dust fires is often a precursor to catastrophic secondary dust explosions. For example, many non-consequential combustible dust related fires occur in dust collectors, duct work, etc. which are non-combustible containers like a chimney flue, boiler, trash compactor, or incinerator.
In the USA fires occurring in non-combustible containers where there are no injuries, fatalities, or property damage is referred as a ‘confined structure” fire. Subsequently, in the National Fire Incident Reporting System (NFIRS 5.0), statistics are compiled so as to understand the probability of occurrence and severity of consequence, which includes near misses. Yet another problem arises in this reporting system similar to the equipment involved in ignition (EII) noted earlier. Most specifically there are no data elements in the NFIRS 5.0 Fire Modules where manufacturing process equipment is considered as “confined structures.” Thus incidents “near misses” occurring in dust collectors or ductwork are not reported or under-reported.
Professional Fire Service Subject Matter Expertise
So how can safety professionals and rulemaking governmental officials effectively evaluate and control future combustible dust incidents? Especially if there is a failure in identifying equipment involved in ignition (EII) or fires in confined structures. This missing link can be rectified through input and subject matter expertise of the professional fire service that responds to fires at manufacturing and industrial facilities.
In a recent Combustible Dust Rulemaking Expert Forum that OSHA held in Washington, D.C. on May 13, 2011, the professional fire service was excluded in the discussion. Instead subject matter experts (SME) comprised organized labor, academia, insurance-underwriters, Chemical Safety Board, and NFPA combustible dust technical committee members being the majority of the fourteen participants. So an organized labor official has more knowledge of combustible dust fires at manufacturing facilities than the fire service that responds to these events?
Connecting the Dots
Acquiring detailed incident data on prior combustible dust fires such as “near misses” where there are no injuries, fatalities, or property damage can assist safety professionals in understanding the location of specific ignition sources and process equipment involved in the combustible dust related fires.
Without following up on this vital information stakeholders will develop faulty reasoning. As mentioned earlier, this is called normalization of deviation, in which nothing catastrophic will occur in the future since nothing bad has happened in the past. A prime example of this reasoning occurred in January 28, 1986 when the Space Shuttle Challenger exploded minutes after the launch. For instance, engineers were aware from prior Shuttle launches with the deficiencies in the O-rings sealing a joint on the solid rocket booster, yet no catastrophe had occurred previously.
In regards to combustible dust, when conducting a process hazard analysis (PHA) in the identification and evaluation of process conditions (equipment), process situations (ignition sources), and process materials (combustible dust) it is extremely important to understand that near misses have a potential for a catastrophic secondary dust explosion.
Combustible Dust Policy Institute
Santa Fe, Texas