Chemical Safety Board Report: Abnormal Situation Management

Step Forward AI

In 2022 the BP-Husky refinery in Toledo, Ohio experienced a large fire in which two men died. In this series of posts we discuss the valuable lessons that we can learn from the Chemical Safety Board (CSB) report to do with that event.

Previous Posts

Posts in this series to date are:

1. Chemical Safety Board Report: Naphtha Release and Fire. An introduction to the CSB’s report on this accident.

2. Chemical Safety Board Report: A Wicked Problem. We note that the refinery had implemented standard hazard analysis procedures and practices. These included Process Hazards Analyses (PHAs), Layers of Protection Analysis (LOPA), and Safety Instrumented Systems (SIS). Nevertheless, two men died. In that post we suggest that new ways of thinking are called for. The Wicked Problem approach is one option.

3. Chemical Safety Board Report: Stop Work Authority. We draw a distinction between ‘Stop Work’ and ‘Stop the Process’.

4. Chemical Safety Board Report: Training or Instrumentation ― Which Is It?

Abnormal Situation Management

The second recommendation from the report reads,

2022-01-I-OH-R2
Revise the Abnormal Situation Management policy to incorporate guidance provided by the ASM Consortium and the Center for Chemical Process Safety (CCPS). The revised policy should include, at a minimum:

a) A broader definition of abnormal situations, such as that defined by the CCPS,

b) Additional predictable abnormal situations and their associated corrective procedures. At a minimum include the following abnormal situations:

1) unplanned crude slate changes,
2) continued operation of the Crude 1 unit with the naphtha hydrotreater unit shut down, and
3) an emergency pressure-relief valve opening.

c) Guidance to determine when an abnormal situation is becoming too difficult to manage and the appropriate actions to take, such as shutting down a process, putting it into a circulation mode, or implementing proper procedures for bringing it to a safe state.

The report defines Abnormal Situation Management as follows,

An abnormal situation is a process disturbance with which the basic process control system cannot cope. Abnormal situations can create a stressful environment for the operators. If abnormal situations are not effectively managed, they can escalate into a more serious incident. In its book, Guidelines for Managing Abnormal Situations, the Center for Chemical Process Safety (CCPS) states: “[s]udden, potentially dangerous situations can affect human performance (the “startle” factor), leading to a “fight or flight” response that can lead to inappropriate action being taken”.

Urgency

The above definition includes the all-important word ‘sudden’.

The majority of abnormal situations do not require an immediate response. Examples of non-emergency situations include:

• The product is off-specification,

• A section of piping is vibrating excessively,

• A piece of equipment is operating badly, but it is continuing to operate,

• Energy consumption has increased by 10%, and no one seems to know why.

In situations such as these, the facility’s managers and supervisors have time to think things through, and to make an orderly decision. However, in crises such as that faced by the operators at BP-Husky time was of the essence ― they had to respond quickly.

Human factors specialists tell us that human error rates rise exponentially during emergencies. In such situations error rates, which are normally in the 1-2% range, can rise to 10-50%. So, let us use an error rate of 25%, and let us assume that the operator has to diagnose what is happening, figure out the best response, and then execute that response ― all in real time, and all with the knowledge that a mistake could result in a fatality.

This sequence of events involves say 10 decisions and actions that need to be taken in sequence. Therefore, the chance of him responding correctly is 0.75^10, or 5%.

A further complicating factor is that one or more of the instruments may be providing incorrect information. (Or even the opposite situation ― I was once involved in a non-emergency situation where an instrument was providing correct information that clearly identified the problem. However, it was ignored because ‘everyone knew that that instrument was no good’.)

Training

One response to these difficulties is to provide more and better training.

While improved training is always desirable, a fundamental difficulty remains. Most abnormal situations such as those faced by the operators at the BP-Husky refinery are unprecedented; they have never been seen before. Hence, it is difficult to prepare a training program that can cover all these contingencies.

The use of simulators can help. The operators can become more adept at thinking situations through in real time. However, it is difficult and expensive to create a simulator for each individual facility.

A Potential Role for Artificial Intelligence

Artificial Intelligence may provide one way of handling abnormal situations more effectively for the following reasons.

• An AI system can monitor large quantities of real-time data, and spot anomalies before they become critical.

• This feature can also help with alarm management.

• The system may be able to identify root causes. This is important on large facilities that have many feedback loops ― both within the process and as part of the instrumentation and controls.

• Having analyzed the situation, the AI system can then test potential responses to see which are most likely to be effective.

• AI is not affected by human emotions, prejudices or bias.

All of the above can be done very quickly.