API RP 754
Process Safety Performance Indicators for the Refining and Petrochemical Industries
|Publication Date:||1 April 2016|
This recommended practice (RP) identifies leading and lagging process safety indicators useful for driving performance improvement. As a framework for measuring activity, status or performance, this document classifies process safety indicators into four tiers of leading and lagging indicators. Tiers 1 and 2 are suitable for nationwide public reporting and Tiers 3 and 4 are intended for internal use at individual facilities. Guidance on methods for development and use of performance indicators is also provided.
NOTE At joint venture sites and tolling operations, the Company should encourage the joint venture or tolling operation to consider applying this RP.
This RP was developed for the refining and petrochemical industries, but may also be applicable to other industries with operating systems and processes where loss of containment has the potential to cause harm (see note). Applicability is not limited to those facilities covered by the OSHA Process Safety Management Standard, 29 CFR 1910.119, or similar national and international regulations.
NOTE To enable consistent application of this RP to other refining and petrochemical industry sub segments, informative annexes have been created to define the Applicability and Process definition for those sub segments. The user would substitute the content of those annexes for the referenced sections of this RP: Annex A-Petroleum Pipeline and Terminal Operation, Annex B-Retail Service Stations, Annex C-Oil and Gas Drilling and Production Operations.
This recommended practice applies to the responsible party. At collocated facilities (e.g. industrial park), this recommended practice applies individually to the responsible parties and not to the facility as a whole. Events associated with the following activities fall outside the scope of this RP and shall not be included in data collection or reporting efforts:
a) releases from transportation pipeline operations outside the control of the responsible party;
b) marine transport operations, except when the vessel is connected or in the process of connecting or disconnecting to the process;
c) truck or rail transport operations, except when the truck or rail car is connected or in the process of connecting or disconnecting to the process, or when the truck or rail car is being used for on-site storage;
NOTE Active staging is not part of connecting or disconnecting to the process; active staging is not considered on-site storage; active staging is part of transportation.
d) vacuum truck operations, except on-site truck loading or discharging operations, or use of the vacuum truck transfer pump;
e) routine emissions from permitted or regulated sources;
NOTE Upset emissions are evaluated as possible Tier 1 or Tier 2 PSEs per Section 5.2 and Section 6.2.
f) office, shop, and warehouse building events (e.g. office fires, spills, personnel injury or illness, etc.);
g) personal safety events (e.g. slips, trips, falls) that are not directly associated with on-site response or exposure to a loss of primary containment (LOPC) event;
h) LOPC events from ancillary equipment not connected to the process (e.g. small sample containers);
i) quality assurance (QA), quality control (QC), and research and development (R&D) laboratories (pilot plants are included);
j) new construction that is positively isolated (e.g. blinded or air gapped) from a process prior to commissioning and prior to the introduction of any process fluids, and that has never been part of a process;
k) retail service stations; and
l) on-site fueling operations of mobile and stationary equipment (e.g. pick-up trucks, diesel generators, and heavy equipment).
Performance indicators identified in this recommended practice are based on the following guiding principles.
- Indicators should drive process safety performance improvement and learning.
- Indicators should be relatively easy to implement and easily understood by all stakeholders (e.g. workers and the public).
- Indicators should be statistically valid at one or more of the following levels: industry, company, and facility. Statistical validity requires a consistent definition, a minimum data set size, a normalization factor, and a relatively consistent reporting pool.
- Indicators should be appropriate for industry, company, or facility level benchmarking.
Process safety incidents are rarely caused by a single catastrophic failure, but rather by multiple events or failures that coincide. This relationship between simultaneous or sequential failures of multiple systems was originally proposed by British psychologist James T. Reason  in 1990 and is illustrated by the "Swiss Cheese Model." In the Swiss Cheese Model, hazards are contained by multiple protective barriers each of which may have weaknesses or "holes." When the holes align, the hazard is released resulting in the potential for harm.
Christopher A. Hart in 2003  represented Reason's model as a set of spinning disks with variable size holes. This representation suggests that the relationship between the hazard and the barriers is dynamic, with the size and type of weakness in each barrier constantly changing, and the alignment of the holes constantly shifting.
Figure 1 depicts both models. In both models, barriers can be active, passive, or administrative/proce