Skip to content

Chemical Safety Board Meeting

Katherine Lemos chair of the Chemical Safety Board

Katherine Lemos, chair of the Chemical Safety Board

On Friday of this week, the Chemical Safety Board will hold an open meeting. Members of the public can call in, but there is no video.

The following is the agenda for the meeting.

The Board will discuss open investigations, the status of audits from the Office of the Inspector General, financial and organizational updates via conference call. The “new business” portion of the meeting will include the possible release of the Kuraray investigation report as well as a discussion led by the Chairman on future plans of the board and how it will be moving forward with a “quorum of one.”

The reference to the “quorum of one” is to do with the fact that the Board currently has just one member, the chair person, Dr. Katherine Lemos. She is new to the position, having been confirmed in March 2020.

The CSB conducts in-depth analyses of incidents at chemical facilities. Their reports are detailed and thorough. They are also often supported by videos illustrating how the incident occurred.

One limitation of the reports is that that they are too limited — they do not provide generalized conclusions that can be used by the process safety community at large. For example, their report to do with the Nitrous Oxide explosion in 2016 at an Airgas facility provides specific recommendations to do with the handling and control of nitrous oxide. These recommendations have little generic value because very few companies actually handle nitrous oxide.

The report does critique the Airgas Management of Change program, but it would be more useful were they to view the incident through a process safety management lens, maybe in tabular form, i.e., to list each of the elements, and to show how they contributed (both good and bad) to the incident.

Safety Moment #58: Fukushima-Daiichi – Two Too Many Common Causes

Common Cause Event at Fukushima-Daiichi

Almost a decade ago, one of the most serious industrial incidents ever to have occurred took place at the Fukushima-Daiichi nuclear power plant in Japan. The event gets little publicity now, even though it is still not known exactly what happened, and even though radioactive materials continue to leak into the surrounding sea water.

One of the themes of the posts at this blog is that we in the process and energy industries can learn a lot from other industries, just as they can learn from us (see Safety Moment #8: “But We’re Different, You Know” in the ebook 52 Process Safety Moments).

The FD event illustrates how supposedly redundant safety systems turn out not be redundant after all. To summarize:

  1. Nuclear power plants must always maintain a flow of cooling water to the reactors following an emergency shutdown.
  2. The 2011 subsea earthquake caused extensive damage to four of the reactors at FD.
  3. The facility went into an emergency shutdown, as designed.
  4. Emergency diesel-powered pumps started and then maintained the flow of water to the reactors. Therefore, even though the damage was extensive, the reactors were in a safe condition.
  5. Then a tsunami that had been created by the subsea earthquake came ashore. It overwhelmed the sea walls protecting the water pumps.
  6. The pumps failed.
  7. The reactors overheated, thus creating a series of meltdowns and fires that released large quantities of radioactive materials into the sea, sky and ground.
  8. The common cause in this event was “subsea quake”. First, it wrecked the reactors, then it destroyed the safety backup systems.

Common Cause Event at Fukushima-Daiichi

Common Cause Event at Fukushima-Daiichi

There are two lessons from this event for those working in the process and energy industries.

  1. Always be on the lookout for difficult-to-spot common causes, and
  2. Recognize that independent safety systems may not actually be fully independent from one another.

Further analysis of this event using Fault Tree methodology is provided in Safety Moment #58: Fukushima-Daiichi – Two Too Many Common Causes.

Safety Moment #106: Process Safety Assessments

Turbine wheel example operating procedures

This week we have released Safety Moment #106: Process Safety Assessments.

All process safety programs have to be audited — typically every three years or so. There is no choice about this — it is usually a legal requirement.

Audits provide valuable feedback to the client’s managers. But they have their limitations. Basically, they generate ‘Yes/No’ answers to statements provided by regulations or company standards. If an auditor offers advice based on his or her personal experience and knowledge, then he or she is no longer an auditor — they have become an assessor.

In this Safety Moment we consider the distinction between process safety audits and process safety assessments.

Audits provide useful insights as to the quality of a facility’s process safety program, and how well it is being implemented. However, those insights are limited, led by experienced industry professionals, are needed to supplement audits.

Grand Canyon Leap — camera does not lie

We have also published Safety Moment #80: Not the Whole Truth.

Also released are Safety Moment #84: Prestartup / Operational Readiness Reviews,

Safety Moment #84: Prestartup / Operational Readiness Reviews

and Safety Moment #59: Risk Perception.

Risk Perception


More Safety Moments

Ebook 52 Process Safety Moments

We have added or republished the following Safety Moments this week.

  • Safety Moment #105: Process Safety Wisdom
  • Safety Moment #87: Hydrogen Sulfide
  • Safety Moment #83: The One-Legged Stool
  • Safety Moment #79: The Process Safety Consultant
  • Safety Moment #77: NORM (Naturally Occurring Radioactive Material)
  • Safety Moment #75: Pressure Tests
  • Safety Moment #67: The Most Important PSM Elements — Management of Change
  • Safety Moment #62: From Complicated to Complex
  • Safety Moment #60: Confined Space Entry
  • Safety Moment #59: In-Kind / Not In-Kind Change
  • Safety Moment #56: Sinking Standards
  • Safety Moment #55: Lifeboat Fall
  • Safety Moment#54: The Process Safety Professional

They are located at

Safety Moment #66: The Most Important PSM Elements — Participation

Offshore oil and gas worker — employee participation

Process Safety Management (PSM) programs are typically organized into a set of management elements. The structure provided by OSHA almost 30 years ago is shown below. (Since then other organizations have developed new systems based on experience, but they all tend to be dialects of the same language.)

  1. Employee Participation
  2. Process Safety Information
  3. Process Hazards Analysis
  4. Operating Procedures
  5. Training
  6. Contractors
  7. Prestartup Safety Review
  8. Mechanical Integrity
  9. Hot Work
  10. Management of Change
  11. Incident Investigation
  12. Emergency Planning and Response
  13. Compliance Audits
  14. Trade Secrets

Each of these elements is important, and they all interact with one another to create a complete system. However, three of them have been highlighted because they are particularly important.

This week’s Safety Moment, Safety Moment #66: The Most Important PSM Elements — Participation, explains why the first of these highlighted elements — Employee Participation — is so important. The Safety Moment stresses that this element is called Employee Participation, not Employee Communication. Nor is it called ‘Culture’.

In future Safety Moments we will consider why Process Hazards Analysis and Management of Change have been placed in the “most important” category.

The New PSM Normal (7) — Things Feel Different

Person thinking about process safety in a changing world

In the post SEMS and Risk in 2020 (SEMS if the offshore equivalent of process safety management as applied to U.S. deepwater operations — mostly the Gulf of Mexico), Mick Will, says,

The speed with which drastic change has come over our industry is what is so different from past events. While oil price volatility has been something I have dealt with for 42 years, I do not remember change happening so fast. I talk to operators who are evaluating what interventions they need to take just to stay in business, with the luxury of long term planning being something they are hoping for in the months to come. This one “feels” different than past ones to me.

I suggest that this feeling of everything being different goes beyond the PSM/SEMS world. In the United States, 30 million people have lost their jobs in just a few months. Even if the virus problem can be solved, we are not going back to Business as Usual; we will not have a V-shaped recovery. In fact, it is quite possible that we will head into a period of prolonged deflation, which, if it continues for long enough, becomes a Depression (see The New PSM Normal (1) — Deflation).

Even some social issues “feel” different. The tragedy of the recent event in Minneapolis has created a response such as we have not seen before. Will that response lead to permanent change? Well, only time will tell, but it does feel different.

When faced with so much change, particularly change that is taking place so quickly, it is easy to become worried and discouraged. Indeed, we do need to be realistic — there are plenty of intractable problems ahead off us. But, when the working environment is changing so drastically, there may be an opportunity to develop new initiatives and better ways of working.

On a much smaller scale, we saw deep shifts in the way that the process industries think about safety about 30 years ago. At that time the chemical industry, in particular, had developed a reputation for being hazardous. Probably the Bhopal event of 1984 was the “highlight” event that generated this feeling. The incident led to the deaths of more than 2,500 people in the local community, and many more serious injuries. (Even if the event was started by a malicious act, the failure of safety systems contributed greatly to the final death toll.) Out of this sense that “something must change” came the widespread development and application of process safety programs in the late 1980s and early 1990s.

Maybe we are at a similar “hinge in time” now — a time when some fundamental changes can be made to the ways in which we manage industrial safety because so much else is changing around us.

The New PSM Normal (6) — Alternative Energy Reality Check

A bright idea for replacing oil reserves

I would like to take a break from talking about this virus this week. (Wouldn’t we all?) Instead, let’s take another long-range look as to where the process industries may be going, and how safety management programs may have to adapt.

There is much talk in the oil industry to do with the current low oil prices and how the loss of revenue will lead to a cutback in investment, something that is already happening. Those oil producers that need high margins, such as the tar sands in Alberta or shale oil in the U.S., are already shutting down much of their capacity. The pandemic (which I promised not to talk about) has led to a further reduction in consumption with a consequent fall in prices.

But, as we discuss in The New PSM Normal (3) — Peak Oil, the fundamental problem that the oil industry faces is to do with the cost side of finding and producing oil. The concept of low ERoEI (Energy Returned on Energy Invested) was introduced in that post. An ever- increasing fraction of the oil that is produced has to be expended on finding and developing new sources of oil. When ERoEI falls below about 5:1 it is questionable if the effort is worthwhile any longer.

A frequent response to these difficulties is that we need to switch to alternative energy sources. The following quotation is from Joe Biden’s current political campaign is an example of this way of thinking. (I am not being partisan here; many people in many walks of life are making similar statements.)

I guarantee you we’re going to end fossil fuel . . . Before 2050, God willing.

The logic is as follows:

  • Fossil fuels (coal, oil and natural gas) generate CO2 and other greenhouse gases when burned.
  • The climate is changing quite quickly as a result and the consequences of those changes look increasingly dire.
  • We have about 30 years to take action before it’s too late.
  • Therefore, let’s use those 30 years to make a transition to alternative energy sources.

All this sounds quite reasonable until we run the numbers.

Here is a back-of-the-spreadsheet calculation, just for the United States. (All numbers are very, very approximate, but they give us a feel as to what can be achieved.)

  • Annual United States energy consumption 1.0E+17 Btu/yr
  • Capacity 3.0E+06 megawatts (MW)
  • Required conversion from fossil fuels to alternative energy over 30 years: 1.1E+05 MW per year.

To summarize, we need to convert about 112,000 MW of capacity every year for the next 30 years just to provide the energy that we use now.

Broadly speaking there are two types of alternative energy: nuclear power and “renewables”, i.e., wind and solar. Let’s say we decide to switch to nuclear power. A representative single reactor plant has a capacity of about 1.7 MW. Therefore we would need to build 67 nuclear power plants a year for each of the next 30 years, starting now. At the end of that 30 years the total number of new plants would be about 2,000. Yet, the actual number of nuclear plants in the U.S. is around 108, and that number is going down — more are being decommissioned than being built.

We can carry out a similar reality check for solar and wind. Currently they provide something like 3% of the nation’s overall energy. Therefore, were we to build out a new energy system built on them we would be effectively starting from scratch. One of the many factors to consider is that now they have the advantage of being the first supplier. When the sun is shining or the wind is blowing, they add energy to the national grid. When it is dark or not windy, conventional power plants pick up the slack. But, if  all of our energy comes renewable sources then we would need enormous backup capability (presumably batteries), or else we would have to keep the conventional plants to be used when needed.

The point of calculations such as these is not to throw cold water on alternative energy projects. They are needed, and they deserve out full support. But they are not going to come close to meeting all of our energy needs in the specified time frame.

What does all this mean for the process industries? It seems likely that the energy needed to run those industries will become increasingly expensive and/or unreliable. But, there is another factor to consider. Something like 10% of a barrel of oil is not burned as fuel; instead it provides the feedstock that is converted into petrochemicals. These in turn are used in the manufacture of the thousands of products that our civilization needs. Therefore, priority should be given to making sure that these feedstocks are available, even if the overall supply of oil is declining.

This type of discussion also highlights the contribution that technical people can make to discussions to do with climate change and other divisive issues. By analyzing ideas and proposals using the basics of math, physics and thermodynamics we can help sort out which of those ideas make the most sense.

The New PSM Normal (5) — Standards

Face mask over a globe symbolizes the COVID-19 pandemic

We are writing a series of posts to do with the ‘New Normal’. They discuss what the world of industrial safety may look like once the current pandemic dies down and business activities resume. Previous posts on this topic are:

In this post I would like to consider how those of us who work in industrial and process safety can help the community at large? The subject came to mind when I was discussing the eventual return of people to church services with a colleague. Our Episcopal diocese has organized a four-phase program for the re-opening of the churches (we are currently in Phase One).

Phases Two and Three require everyone to wear a face mask while on church premises. But the guidelines do not specify what type of mask. Right now, due to the unavailability of commercially-made items, many people are wearing home-made cloth masks. These are good for protecting other people, and they set the right tone — people who wear masks are sending a message that they care. However, there is no control over the design or effectiveness of these masks. Any type of mask offers substantial benefits, so they should always be worn. However, the effectiveness of the individual masks is bound to be quite variable.

I suggested to the church leaders that we should specify that masks should meet an industrial standard such as ASTM Level 2, as shown in the following chart.

ASTM Standards for Face Masks

The suggestion was not taken up. Maybe the reason for the pushback was that most people are not used to a culture of safety standards. Yet such a culture is taken for granted by readers of posts such as this. Were one of us to show up at a chemical plant or refinery wearing a home-made hard hat we would be shown the exit immediately.

So part of the ‘New Normal’ may be for safety professionals to urge the use of standards such as those provided by ASTM when it comes to responding to the crisis.

ASTM standards for use in COVID-19 pandemic

The following information was taken from the ASTM web site.

ASTM International is providing no-cost public access to important ASTM standards used in the production and testing of personal protective equipment – including face masks, medical gowns, gloves, and hand sanitizers – to support manufacturers, test labs, health care professionals, and the general public as they respond to the global COVID-19 public health emergency.


The New PSM Normal (4) — Automation

Dog offshore platform

If you fly over the offshore platform of the future in a helicopter and look down you will see just two living beings: an operator and a dog. The operator’s job is to feed the dog; the dog’s job is to make sure that the operator doesn’t touch anything.

This is the fourth post in our series to do with the ‘New Normal’ as it applies to industrial safety. The series is written on the assumption that the current pandemic is so severe and so sudden that we cannot go back to the ‘Old Normal’. All aspects of our lives, including the way in which we manage safety, will change.

Of course, none of us know what the future holds. We are still in the middle of this pandemic; it is still growing and there seems to be no end in sight. Still, now is the time to consider what the future may hold and what opportunities may present themselves.

Only a crisis – actual or perceived – produces real change. When that crisis occurs, the actions that are taken depend on the ideas that are lying around. That, I believe, is our basic function: to develop alternatives to existing policies, to keep them alive and available until the politically impossible becomes the politically inevitable.

Milton Friedman

Some of the ideas that are “lying around” have been discussed in previous posts.

  • In The New PSM Normal (1) — Deflation we suggest that our economy is entering a time of deflation, which is defined as a situation where there are more goods and services available than there are people with money to purchase them.
  • In The New PSM Normal (2) — Do Less With More we consider the vulnerability of our “lean and mean” supply chains — a vulnerability that the virus has exposed. An effective industrial policy may be to install lots of spare equipment and redundancy to take care of surprises.
  • Although the pandemic is a crisis that is up front and center, we may find that the changes that are taking place in the oil industry will eventually have a greater impact. There has been considerable discussion to do with the low price of oil. What gets less attention, but which is actually more important, is the cost of finding and developing new sources of oil. This is an important but complex issue that will be discussed in a series of posts, starting with The New PSM Normal (3) — Peak Oil.

In this post we start to think think about automation. A steady and consistent trend in all industries over recent decades has been the installation of sophisticated control systems that provide tighter control of the processes that they are operating than can be achieved by human operators. These systems are also less costly than skill labor.

As we emerge from the pandemic we are likely to see two conflicting points of view. The first is that the economic incentives for installing these systems have not changed. Indeed, one advantage of an electronic systems is that they do not need to keep a social distance from one another. With regard to employment, it will be argued that efficient systems grow the overall business, and so allow for people to be employed in new areas.

On the other hand, with over 30 million people unemployed in the United States alone, there will be intense political pressure to put people to work, right now.

For those of you who have an interest in the ‘New Normal’ for people of faith, this week’s post is Pontius Pilate and Truth.

SEMS Audits


BSSE (the Bureau of Safety and Environmental Enforcement) has published an overview of “Common Operator Deficiencies Discovered by Safety Environmental Management System (SEMS) Audits”. It is Safety Alert No. 381 at