I.CHEM.E. SYMPOSIUM SERIES NO. 110

MANAGING SAFETY IN THE CHEMICAL INDUSTRY

P. L. Thibaut Brian*

All accidents are preventable, and safety can be managed. Managing safety requires molding corporate culture as well as developing superior technology. It requires building proper attitudes and safety awareness as well as conducting hazard reviews and safety audits. A total safety program is required to prevent falls from ladders and runaway reactors, cuts and burns and explosions and toxic releases. The leadership must come from the chief executive officer. He must make safety his top corporate priority.

Safety is a very serious responsibility of chemical industry management and of the chemical engineering profession. Ours is a hazardous business -processing, storing, and transporting huge volumes of materials which are flammable, toxic, and explosive. Our highest priority in the conduct of that business must be the protection of our employees, our customers, our neighbors and the general public. This is an enormous managerial and tech­nological challenge. Moving a corporation to a leadership position in chemical industry safety requires molding the culture of the entire organi­zation. It requires penetrating attention to detail in design, operation, and maintenance. It involves motivating and training everyone in the entire corporation to always put safety first - to always ask what could possibly go wrong and to never take shortcuts. It requires diligent investigation of all accidents and near misses, with no tolerance of cover-ups. People do not readily accommodate to these disciplines. They require the strongest kind of leadership from the top.

This leadership role must be assumed by the Chief Executive Officer of the corporation. He is responsible for the safety of its employees and for the impact of its operations on the safety of the general public. He must make safety his highest priority. He must demand superior safety performance of his entire organization. He must require his management team to set meaningful safety goals, to measure safety performance, to reward superior performance, and to refuse to tolerate mediocrity. He must demand a total safety effort, including all aspects of safety - from falls off ladders to runaway reactors. It might be tempting to think that we can achieve a high degree of safety with respect to major accidents which affect the general public while ignoring, in a relative sense, worker injuries resulting from

* Air Products and Chemicals, Inc., Allentown, Pennsylvania, U.S.A.

615

I.CHEM.E. SYMPOSIUM SERIES NO. 110

t r i ps and f a l l s and cuts and burns. I am convinced that th is is not achievable. There is a correlat ion between the more frequent minor accidents and the less frequent major accidents. There is a correlat ion between process safety and hard hat safety. The common denominator i s the people and the i r d isc ip l ine and the i r a t t i tudes. Inattent ion to hazards, short cuts and r isk taking, and sloppy habits - these a t t i tud ina l deficiencies resu l t in major as well as minor accidents.

Safety i s a management respons ib i l i t y , and safety can be managed. Industr ia l accidents do not j u s t happen. A l l accidents are preventable. Management has the responsib i l i ty and the a b i l i t y to control industr ia l accidents. As evidence to support th is contention, Figure 1 shows the lost- t ime in jury frequency rates of a number of chemical companies in the United States as reported to the Chemical Manufacturers Association to compete for the Lammot DuPont Safety Award. The lost- t ime in jury frequency rate is the percentage of the workers in the corporation who suffered in a given year an in jury so severe that i t caused the worker to miss some work­days. Back in 1974 the lost- t ime in jury frequency rate for Air Products was 1.6, approximately equal to the average of the companies reporting to the Chemical Manufacturers Association. Our Chief Executive Off icer was not sa t i s f ied with an average safety performance, and he challenged his manage­ment team to move Air Products into a leadership role in safety. By 1985, the lost- t ime in jury frequency rate for Air Products had been reduced to 0.04, a f o r t y - f o l d reduction in eleven years. The CMA average declined over that period, but the decline was not nearly so great. DuPont, the industry leader, had a very enviable in jury rate of 0.2 in 1974, and they were able to gain a reduction of a factor of ten, giving them a rate of 0.02 in 1985. Look at the performance of Dow chemical. Their lost- t ime in jury rate was reduced from 2.8 in 1975 to 0.01 in 1985 - a two hundred e ighty- fo ld reduc­t ion in ten years. These dramatic improvements did not happen by accident! They happened because the management of the company, from the Chief Executive Off icer on down, became to ta l l y dedicated to operating in a f i r s t -c lass, safe manner. Shown without names are the records of several other companies whose performance is better than average but who showed l i t t l e or no improvement in th i s decade. Not shown on the s l ide are the curves for companies with in jury rates greater than the CMA average. The in jury rates of those companies range up to values between 5 and 10 - f ive hundred-fold higher than the in jury rates of Dow and DuPont. No one could believe that the operations of those companies are inherently f i ve hundred-fold more r isky than the operations of Dow and DuPont, but the i r safety management programs are surely f i ve hundred-fold less e f fec t ive . Safety rea l ly can be managed.

Managing safety is j us t l i ke managing sales, p r o f i t s , product iv i ty , or costs. The i n i t i a t i v e must come from the top. When the Chief Executive Off icer meets with his senior management, i f he g r i l l s them in great detai l regarding sales, p r o f i t s , costs and ef f ic iency and then as they walk out of the door he casually asks them "By the way, how is the safety of your opera­t ion going?", you can rest assured that safety performance w i l l be mediocre. He must penetrate into the safety record of each of his operating units as intensely as he does into overhead costs.

A very e f fec t ive technique is for the Chief Executive Of f icer to i ns t i t u t e a policy of having safety as the f i r s t item on the agenda of his senior management committee meetings. Any serious incident throughout the corpora­t ion should be reported promptly, and the senior l ine o f f i cer of the organi­zation involved should make a presentation on what happened, why i t

616

I.CHEM.E. SYMPOSIUM SERIES NO. 110

happened, and what is his plan of corrective action to prevent recurrence. Safety must be understood and accepted to be a line management respon­sibility. Staff safety professionals can support line management by developing safety programs, but line management must, accept the respon­sibility for safety. Senior management should periodically review the statistics on worker injury rates, not just for the entire corporation but also for the major divisions. Inevitably, the safety performance of the best division will be found to be far superior to that of the worst, and an open review of this fact develops considerable peer pressure for improvement.

There is nothing easy about remolding the culture of a corporation, but that is what is required. In the past if an Air Products worker caused his own injury, his supervisor and higher management often took the attitude "I told him to work safely but he did not do it". Such an attitude is no longer a part of our culture. Today, a worker injury is recognized to be a manage­ment failure. Management must install the programs to prevent accidents, and these include worker training and motivation and the discharge of workers who will not cooperate. Back in 1974, when we began our program, we heard many excuses from people who were slow to accept this new philosophy. You can probably visualize the various arguments we heard on why we should not compare the safety record of an industrial gas business with that of a chemical business, or a construction activity with an operating activity, or a European operation with an American operation. None of the arguments we heard were completely without merit, but they all missed the main point. Safety achievement is, in fact, far more dependent upon management commit­ment and attention than upon the type of industrial activity, the geographic location, or all of the other factors so often invoked to attempt to justify mediocre performance.

Following from management commitment and attention, many safety programs must be developed in order to insure that designs are reviewed and checked, operations are documented, reviewed and checked, people are trained and retrained and motivated, and safety audits are vigorously pursued to ferret out all potential accidents just waiting to happen. In our industry, a strong engineering safety effort must be a vital part of this program.

At Air Products, we have placed great emphasis on the use of formal hazard reviews of all of our operating facilities and all new plant designs. We have educated 1,700 of our personnel in the techniques of hazard analysis, and more than half of them have participated in formal hazard reviews. We believe it is beneficial to involve a large number of our people in our hazard review efforts. While most of them will participate in a formal hazard review only occasionally, this training and experience will, we believe, help them to think logically about process safety in their every day work. We have also placed strong emphasis upon quantification of the probabilities of our major hazards, and more than 150 of our personnel have received formal education in this area. We now have a policy of quantifying the probabilities of all of the major hazards which we can quantify with reasonable confidence. We were compelled to follow this course by the shock we experienced when we quantified the probabilities of several of our major hazards and found them to be far more probable than we had perceived them to be based upon qualitative assessments.

The decision to quantify risks brings with it another challenge - one must develop a method for deciding how safe is safe enough. Once a risk is quantified, it is usually possible to propose ways of reducing it. The risk can never be reduced to zero, but it can usually be reduced further,

617

I.CHEM.E. SYMPOSIUM SERIES NO. 110

inevitably with the expenditure of more money. One is forced to decide where to stop. Our approach to making th is decision is based upon the comparison of the r isk in question with other r isks . I f the r isk is large compared with the other r isks involved in running our business, i t is imperative that i t be reduced. I f the r isk is quite small compared to the other r isks involved in our operations, a decision is made to accept i t for now and to devote our resources to the reduction of our larger " isks . But decisions such as these can be made only by the senior managers •". of the corporat ion, and I believe that senior management must be involved int imately in th is aspect of the engineering safety program.

Needless to say, r isk quant i f icat ion must always be applied with good judgment, and i t should be viewed as a supplement to , not a substi tute f o r , engineering design standards, best industry pract ices, and safe designs based upon long experience. Some companies may not f ind r isk quant i f icat ion to be suited to the i r needs, and I do not recommend i t for everyone. I w i l l say, however, that our company has found r isk quant i f icat ion to be extremely useful as a means of uncovering r isks which are far greater than had been perceived and as a means of ident i fy ing the most ef fect ive ways to reduce those r i sks .

Some may question whether we can afford the cost of elevating safety to our highest corporate p r i o r i t y . I am convinced that the benefits far exceed the costs. In addition to the reduction in worker in jury rates, our company has experienced lower costs for workers' compensation insurance, damage to plant and equipment, los t production, and l i a b i l i t y claims. We have also exper­ienced improved worker morale, better product iv i ty , and improved workmanship qual i ty . We are convinced that a f i r s t - c l ass safety management system is a necessary ingredient in a f i r s t - c lass management system.

In closing I wish to return to a v i t a l point . An elaborate engineering safety e f f o r t without a tota l safety program is not adequate, in my judgment, to guard against major process accidents which can impact the general publ ic. The engineering department might design the world's safest plant with the world's best safety protection systems, but i f the people operating that plant and the people maintaining that plant are not highly trained and highly motivated to always do the i r jobs in a f i r s t - c l ass manner, to always put safety f i r s t , to always question what might possibly go wrong, and to never take shortcuts, that plant operation w i l l not achieve the level of safety to which we aspire. This safety t ra in ing and motivation involves heightening the safety awareness of everyone in the corporation. I t involves modifying att i tudes and always putt ing safety f i r s t , in the o f f ice and in the plant, in the control room and on the highway, at work and also at home. People cannot turn on the proper safety at t i tude when they come to work and then turn i t of f when they leave for home. They cannot turn i t on when maintaining the pump but turn i t o f f when securing the ladder. They cannot turn i t on when operating the reactor but turn i t o f f when dr iv ing the t ruck. Proper safety at t i tudes can be developed to the level at which they are needed only by to ta l safety e f fo r ts which teach and emphasize and reemphasize doing things r igh t - so that we avoid cut f ingers and broken bones and major explosions and toxic releases. That is the level of safety that society expects of us, and we can set t le for nothing less.

618

I.CHEM.E. SYMPOSIUM SERIES NO. 110

619