Working Party Lessons from Process Safety Incidents

Chair : Dr. P. G. Schmelzer, Bayer HealthCare AG, Leverkusen

I. Objectives

"Learning is like swimming against the stream. If you stop swimming, you will drive backwards."
Lao Tse (300 before Christi)

The accident statistics show the excellent level of safety engineering within German Chemical Industry, but even one big accident could challenge this statement effectively, at least to the public perception. The Working Party supports learning, particularly from non-notifiable incidents relevant to safety. It aims to provide factual information that can be widely applied in practice pragmatically and unbureaucratically, and to constantly update the pertinent theory through publications.

This procedure has been implemented since 1996 by a group of experts as Working Party ‘Lessons from Process Safety Incidents’, who analyzes incident descriptions, submitted by VCI members. These are first anonymised, then analysed and, provided they are of sound educational value, they are made available to all who are interested in standardised form. This is realised by means of a publicly accessible database on the Internet (http://processnet.org/en/incident_db.html).

Proceeding:
VCI member companies

1. systematically collect data of non-notifiable (under German legislation) incidents relevant to safety
2. provide a summary in the company-specific manner, and
3. send this summary together with additional information (if applicable) to the person in charge at ProcessNet
4. The person in charge of ProcessNet anonymises the all information and distributes it to the members of the Working Party' Lessons from Process Safety Incidents'
5. The Working Party, consisting of experts from science, authorities and industry evaluates and reviews the material
6. The evaluated, revised (if applicable), and anonymous summary is published in the incident database

The online database offers five descriptors to support your search:

1. Basic operation / process
2. Plant part affected
3. hazard features / substance properties
4. cause / effect
5. other (operating method, organization, environment, etc.)

Additionally you can search by free text.

II. Interfaces to other committees

The assessment of industrial safety related incidents involves several disciplines. Thus you might have overlapping with the following ProcessNet Subject Divisions (FG):

• FG Chemical reaction technology
• FG continuing education (knowledge transfer)
• FG Process, instruments and equipment technology, as well as
• all Working Parties of the FG safety technology.

The Working Party continously enrich the database with new incidents.

III. Key activities

The following issues are addressed in several incident descriptions / summaries:

1. Safe separating of piping and appliances, recognition of ex-zones
2. Avoid incorrect ranges of temperature-/pressure-/concentration
3. Compatibility of substances, additives and raw materials used
4. Discarding of residues/decompression of piping and appliances
5. Ignition source due to mechanical friction/dissipated energy, keeping of ex-zones
6. Product release at sampling and sealing systems
7. In-service inspections and maintenance of components, especially at critical structural parts/procedure bodies
8. Hazard of substance accumulation by plugging/deposition
9. Efficiency of organizational safety provisions by adequate documentation and technical measures
10. Short-term breaks of operation or power supply
11. Electrostatic discharge as ignition source, keeping of ex-zones
12. Safety consideration in preparation for the safe handling and inspection of repair work and for changes to equipment and their use/operation mode

The incident summaries compiled so far and the lessons learned can be summarized as the known, but important following conclusions:

1. With the help of systematic safety provisions and the consequent use of common knowledge most of the incidents can be avoided.
2. Profound knowledge about the behavior of substances has to be obtained, even beyond the normal rangesof process parameter (T, p, c etc.)
3. Also issues deemed safe have to be continuously challenged with respect to their efficiency regarding potential hazards and their reliability in the field.
4. Processes, equipment and workflows should be designed and organized as simple as possible.
   

IV. R&D-key aspects

The Working Party does not necessarily convey a R&D-demand from its work.

April 2007