MeetingsRoundtable Meeting — June 2-3, 1999Benchmarking Process Safety Programs Briefing Paper No. 4 Mary Kay O’Connor Process
Safety Center Roundtable Meeting John Noronha
Table of Contents
I. BACKGROUND OF OVERALL CHEMICAL SAFETY ASSESSMENT PROGRAM The benchmarking project is an integral part of the program, and will be used in various phases of the program. So we will describe the background of the overall program. Later we will focus on some benchmarking aspects. The objectives of the overall chemical safety assessment program are very important to the public, industry and the government. They will impact, not only safety issues, but also on business being more competitive (Fig. 1). The Stakeholders and the Regulations noted in Fig. 1 are shown in Fig. 2 & 3. Special attention should be paid to the last two regulations in Fig. 3, namely; OSHA, PSM, and EPA RMP. They represent a unique turn for regulatory agencies. Both were mandated to be developed under the Clean Air Act amendments and can be considered unique because they essentially require affected sites to establish a proactive business philosophy toward preventing and responding to chemical releases. They do this by requiring management systems to be developed. Neither regulation has many hard and fast detailed requirements. The benefit to a plant site comes from avoiding the damage that may result from an accidental release. The benefits to the public living or working in the vulnerability zone include a stronger sense of security about the chemical plant that may be a neighbor. The project team (Fig. 4) and case history basis (Fig. 5) are vital aspects of this chemical safety assessment program. The 66,000 reporting facilities under RMP are generally classified under 10 categories noted in Fig. 6. The program will be done in 4 phases (Fig. 7) to establish different segments of the program and to allow reflection on how and whether to proceed to the next phase. The four phases are:
At this June 2 & 3, 1999 meeting, five white papers (Fig. 12) will be presented and discussed with about 50 representatives from the public, industry and government. They will relate primarily to Phases I and II, but also touch on Phases III and IV. They are:
II BENCHMARKING PROJECT PRIMARY FOCUS: CAN BENCHMARKING LEAD TO OBJECTIVE RESULTS? As noted earlier, benchmarking is an integral part of the entire program, and it is intended to be used in various phases of the overall program. The primary focus is to evaluate whether we can use benchmarking objectively to assess chemical safety and risk management programs, including technologies and subprograms (Fig. 13). Two specific questions are:
If the answer is yes to both these questions, then how
would the work progress? What kind of data would we have to collect? Where
would the data come from? And how would the whole issue of benchmarking
be handled? III A) Regulatory and Industrial Chemical Safety Programs to be Evaluated in Benchmarking For benchmarking, the major regulatory chemical safety programs relating to toxic, reactive, flammable and explosive of accidental releases are shown in Figure 3. For industrial safety programs, they are:
III B) Stakeholders Used in Benchmarking In benchmarking, there are many types of stakeholders involved in the effort to reduce the risk of toxic, flammable and explosive substances due to accidental releases as shown in Fig. 2. They are regulatory, state and local governments, industry, emergency response organizations, environmental groups and the public. III C) Benchmarking for Goals/Objectives of Stakeholders 1) Determining National Safety Goals The goals and objectives of the overall chemical safety assessment program and of stakeholders are shown in Fig. 1, and discussed earlier in this report. Different stakeholders will likely have different goals and objectives? Some examples of goals are as follows: How many deaths and injuries would be acceptable in a given period of time? How much property damage and business down time is acceptable? How many releases and how large a release is acceptable? How much time and money can be spent on hazard analysis? Our project goal is to get benchmarking data from a representative sample of stakeholders (Fig. 2) to determine their goals and objectives. Then identify commonalties and differences, and then identify what we will consider national chemical safety goals (Fig. 9). III D) Potential Benchmarking Reporting Facilities Our benchmarking efforts could be based on (Fig. 5):
III E) Potential Accidental Release Data There are several sources of accidental release data (Fig. 8) for benchmarking. A key reference (Ref. 4) is entitled, "Accident History Database: An Opportunity, "(M. Sam Mannan and T. Michael O’Connor and Harry West) accepted for publication in Environmental Progress. Here are some examples:
LEPC’s have program requirements (Fig. 14) which relate to various safety programs used and the 5-year accident history. Program 1 is the least stringent RMP for "lower hazard" processes. A process qualifies for Program 1 if:
A process is in Program 3 if it does not qualify for Program 1 and it is either:
Details of the 5-year history are shown in Ref. 2, page 26. They relate to both on-site and off-site deaths, injuries, property (see also Fig. 8) on Phase I - Data and Analysis of Accidental Releases. 3) Company Plant Accident Reports Companies are required to report accidents due to government regulations, and often because of internal company policies. Appendix 2 (Ref. 2, pages 249-252) entitled "Example of plant Accident/Incident Reporting Procedures" describes the details, the purpose, scope, policies, procedures, for both the Safety Department and Line Departments; and the Guidelines for conducting a formal investigation of accidents. 4) Industry Group Surveys Another potential data source is the recently initiated CCPS Incident Database program. It reviews how data is collected and analyzed for about 25 companies who volunteered and paid dues to be part of the program and is only available to those companies. The proposed 1999 DIERS benchmarking project (Appendix 4) is yet another potential data source of benchmarking on how engineering systems are designed and analyzed to prevent and mitigate accidental releases from exothermic runaway and decomposition reaction systems. It is based on a paper entitled, "Why DIERS Technology Should be used in Risk Assessment: Call for 1999 Worldwide Benchmarking Survey on Various Risk Reduction Methods Used." (Ref. 5). III F) Developing Metrics and Measurements A key element of the overall program is to evaluate whether we can, in fact, use benchmarking to assess chemical safety programs including technologies, and subprograms (Fig. 13). To do so, we need to develop methods for metrics and measurements (Fig. 10). We would need to analyze data by the identification of general normalized measures, including qualitative and quantitative measures. Appendix 5 is an example of plant follow-up procedures for accidental investigation. (Ref. 2., pages 249-252). Ref. 4 is a good reference of "accountability" objectives and goals of process safety as a management function. (Ref. 3, Chapter 3, pages 15 to 23). As shown in Figs. 8 to 10, we need to identify factors that are responsible for trends. We could use Delphi Technique on companies that are likely to comply with safety programs. We could use confidential interviews for companies (that are not complying) to determine if their safety records are statistically worse, better, or the same. We need to analyze whether only subjective conclusions are possible (rather than objective conclusions.) from the proposed benchmarking. III G) Develop Report Finally, we would need to develop a report to meet the objectives of the chemical safety assessment program. Then benchmarking effort and related efforts discussed would be key to its success. A general outline of the report is described in Fig. 11. IV ADDITIONAL WORK TO IMPROVE THIS DRAFT John Noronha will present this as a white paper at this June 2-3,1999 meeting of key members of the program and key stakeholders. This is in preparation for the October 27-28 conference, and ultimately for the actual benchmarking project work and the final report on the overall program. APPENDICES
Figure 1: OBJECTIVES
Figure 2: STAKEHOLDERS
Figure 3: SOME KEY REGULATIONS
Figure 4: PROJECT TEAM
Figure 5: CASE HISTORY BASIS *Case Histories and Studies Could Be Based on
Figure 6: TEN EPA SELECTED NORTH AMERICAN INDUSTRIAL CLASSIFICATION SYSTEM (NAICS) CODES
Figure 7: PHASED APPROACH Purposes: 1) Establish Different Segments of Programs Four Phases: I. Describe Evolution of Chemical Safety II. Describe Stakeholder's Goals and Objectives for Improving Chemical SAFETY III. Develop Indicators, Measures and Metrics to Measure Progress Towards Goals and Objectives IV. Develop Report Figure 8: PHASE I DESCRIBE EVOLUTION OF CHEMICAL SAFETY Page 1 of 2 Focus: Assessment of all available data2 and analysis techniques3, for accidental release rates1 1 Accidental release rates in terms of number of:
Figure 8: PHASE I DESCRIBE EVOLUTION OF CHEMICAL SAFETY Page 2 of 2 2 Data Sources Include:
3Analysis Will Be Identification of General Normalized Measures, Including Qualitative and Quantitative Measures (As Available)
Figure 9: PHASE II DETERMINE/DESCRIBE STAKEHOLDERS GOALS AND OBJECTIVES FOR IMPROVING CHEMICAL SAFETY Goal: Determine What They Are? How to do? General: Study the Phase 1 Information and Applying Assessment Techniques Specifically: 1) Identify Stakeholders (Fig. 2) 2) Identify Their Goals and Objectives 3) Identify the Commonalties and Differences of Goals 4) Identify National Chemical Safety Goals
Figure 10: PHASE III DEVELOP INDICATORS, MEASURES, AND METRICS TO MEASURE PROGRESS TOWARDS GOALS Page 1 of 2 Goal: Identification of Programs and/or Other Factors that are Responsible for Trends How to do? 1) Delphi Technique on Chemical Companies (Because They are More Likely to Comply) 2) Confidential Interviews will be Made with Companies (that are not complying) to Determine if Their Safety Records are Worse (Better) or Statistically Different 3) Analyze Whether Only Subjective Conclusions are Possible (Rather Than Objective Conclusions) Figure 10: PHASE III DEVELOP INDICATORS, MEASURES, AND METRICS TO MEASURE PROGRESS TOWARDS GOALS Page 2 of 2 Specifically: 1) Identify Measures/Indications for National Chemical Safety Goals 2) Identify Activities (Of Stakeholder Programs) That Contribute To Indicators And Data Collection From Stakeholders For Measures (Ref. 3, page 5) 3) Determine Cause/Effect Relationship Between Stakeholders Goals And Activities
4) Identify Gaps in Meeting National Chemical Safety Goals *5) Identify Ineffective Activities Which Should Be Improved Or Eliminated 6) Develop Metrics to Measure Progress Towards National Chemical Safety Goals Figure 11: PHASE IV DEVELOP REPORT To Include:
- Detailed description of metric, and application of metrics in measuring progress towards meeting national chemical safety goals. *2) Identify programs that should be emphasized, enhanced, modified or dropped (AIChE, API, ASME, ASTM, CMA, DIERS, NFPA, PSM, AND RMP). 3) Identify what trends in the industry may lead to future problems or improvements (for example, does TRI Reporting cause companies to take actions that result in process safety improvements). 4) List concepts on how we could reduce or eliminate some safety requirements in the design, production, maintenance, and training (so as to make business more competitive while still operating at generally acceptable risks).
Figure 12: FIVE WHITE PAPERS FOR JUNE 2-3, 1999 MEETING 1. Overview of Chemical Safety (Dr. Irv Rosenthal/Dr. Sam Mannan) 2. History of Chemical Safety and Risk Management. (Jon Averbach) 3. What Do Various Accident Databases Tell Us (Fig. 8)? (Eboni McGray) 4. Use of Benchmarking to Assess Chemical Safety and Risk Management Programs. (John Noronha) 5. Insurance Perspective on Future Direction of Chemical Safety and Risk Management Programs. (Paris Stavrianidis)
Figure 13: BENCHMARKING PROPOSAL FOCUS: The Ability to Use Benchmarking to Assess Chemical Safety and Risk Management Programs, Including Technologies and Subprograms Question: Can We Benchmark to: Q 1) Understand Where We are in Comparison to National Chemical Safety Goals (if we are able to establish these elusive goals) (Fig. 9)? or Q 2) Develop Metrics and Measurement Systems for (Measuring) Progress Towards National Chemical Safety Goals (Ref. 2, pages 246-250)? If Answer is Yes to Q1 and Q2, Then,
Figure 14: LOCAL EMERGENCY PLANNING COMMITTEE GUIDEBOOK Program Requirements Summary
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
About
NCSP | Reports | Data
| Tips | News
| Calendar | Education
| Contact us Texas
Engineering Experiment Station
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||