Showing papers in "Journal of Loss Prevention in The Process Industries in 1994"

Employee attitudes and safety in the chemical industry

Abstract: In recent years, there has been growing discussion about the role of human factors in safety, beyond the basic level of errors and slips. This discussion has focused on social psychological issues such as safety attitudes and safety climate. To date, only limited research has attempted to measure these factors and to establish empirically whether and to what degree they are related to safety performance. This paper describes the development of a question set and a set of scales to measure safety attitudes in the chemical industry. A question set comprising ten scales was developed, and used to survey ten chemical sites. The results show that the scales are reliable measures of safety attitude. Further, correlations between the companies' scale scores and accident rates revealed a statistically significant relationship. The results are discussed together with directions for future research.

The boiling liquid expanding vapour explosion

Abstract: The boiling liquid expanding vapour explosion (BLEVE) has existed for a long time and for most of this time it has been cloaked in mystery. Several theories have been put forward to explain this very energetic event but none have been proven. This paper describes a series of tests that have recently been conducted to study this phenomenon. The study involved ASME code automotive propane tanks with nominal capacities of 400 litres. The tanks were exposed to a combination of pool and/or torch fires. These fire conditions led to thermal ruptures, and in some cases these ruptures resulted in BLEVEs. The variables in the tests were the pressure-relief valve setting, the tank wall thickness, and the fire condition. In total, 30 tests have been conducted, of which 22 resulted in thermal ruptures. Of those tanks that ruptured, 11 resulted in what we call BLEVEs. In this paper, we have defined a BLEVE as the explosive release of expanding vapour and boiling liquid following a catastrophic tank failure. Non-BLEVEs involved tanks that ruptured but which only resulted in a prolonged jet release. The objective of this study was to investigate why certain tank ruptures lead to a BLEVE rather than a more benign jet-type release. Data are presented to show how wall temperature, wall thickness, liquid temperature and fill level contribute to the BLEVE process.

Combustion hazards posed by the pressurized atomization of high-flashpoint liquids

Abstract: Study of the revised Institute of Petroleum Model Code of Safe Practice for area classification of petroleum installations has raised the question of whether pressurized systems of high-flashpoint liquids (e.g. diesel, kerosene) should be considered hazardous in respect of their potential to atomize to produce flammable aerosol systems. This work collates relevant scientific studies to assist in investigation of the possibility of combustion hazards posed by these systems. It is envisaged that it will serve as an initial assessment in an area where industry acknowledges that insufficient information exists. A simple method is presented for characterizing liquid leaks in terms of their potential to atomize, which is considered a necessary condition to give rise to a flammable atmosphere. A cloud of very finely atomized liquid near stoichiometric concentration is shown to be hazardous with regard to a ‘low-energy’ spark. Empirical formulae for estimating the initial ‘average’ droplet size and the functional form of the volume of fuel contained in droplets that are less than a particular droplet size for this type of leak give an indication of the nature and potential severity of the hazard. These correlations need to be validated for use outside their ranges of applicability or to be updated. Information collated here facilitates what is likely to be a very conservative assessment of the hazard potential. Identification of areas where there is a dearth of information is just as important, and it is hoped that this paper will provoke further interest leading to improved methodologies for hazard quantification. However, at this stage, it has to be concluded that, even under low pressures, the possibility of creating aerosols that result in flammable systems cannot be eliminated for commonly-handled high-flashpoint liquids. Two practical illustrations, a large-scale study and a real incident, are given to support this conclusion. Finally, in the specific context of area classification, a simple solution to some potentially hazardous aerosol leaks is suggested.

Uncertainty representation and propagation in quantified risk assessment using fuzzy sets

Abstract: It is generally acknowledged that there are substantial uncertainties present in any analysis of risk. This paper provides a brief overview of the current techniques used for uncertainty analyses, and highlights their inappropriateness for practical use in the complete risk assessment process. The concept of fuzzy sets as a means for quantifying uncertainty is introduced and a case study demonstrates the application of this method to a simple consequence analysis where parameter uncertainty is considered. The results of this fuzzy analysis are compared with those of a more traditional probabilistic approach using a Monte Carlo simulation. This comparison demonstrates that the novel approach of fuzzy sets is a more appropriate technique due to its non-statistical nature and that the amount of computation required is substantially reduced compared to the traditional probabilistic approach. The versatility of fuzzy set theory suggests that this approach could also be used to quantify other types of uncertainty present in the risk assessment process, including model uncertainty and expert opinion.

Possible sources of ignition of potential explosive gas atmospheres on offshore process installations

Abstract: This paper highlights the basic features of various ignition sources and some principles for prevention of ignition of potential explosible hydrocarbon gas atmospheres in industrial practice. Ignition by hot solid surfaces, heat from mechanical impact and friction, electric discharges and jets of hot gaseous combustion products is discussed in detail. The possibilities of effective ignition sources in industrial practice being generated by stray electric currents, cathodic corrosion protection currents, static electricity, lightning, high-frequency electromagnetic waves, optical and radioactive radiation, ultrasonic radiation and heating by compression are discussed.

Quantitative risk assessment applied to offshore process installations. Challenges after the piper alpha disaster

Abstract: The Piper Alpha disaster in 1988 has led to widespread adoption of quantitative risk assessment (QRA) in decision-support within the North Sea oil and gas industry. This has presented a challenge to practitioners in QRA, both with respect to the reality and detail which their predictions can represent and in terms of the speed and traceability of their work. This paper uses examples to illustrate the extensive advances which have been required in offshore QRA during the past five years. It traces the growth of time-dependent consequence modelling for leaks from risers and process equipment, and the adoption of linked spreadsheets to speed up the analysis of escalation risks for a full range of such events across a large platform. The quality assurance challenges relating to such complex spreadsheets prompted the development of the multi-sponsored Offshore Hazard and Risk Analysis (OHRA) Toolkit as an industry standard for QRA applied to offshore installations.

Mitigation of problem clouds

Abstract: This paper describes the use of fluid curtains for the mitigation of problem clouds. Mechanical and physico-chemical actions of spray barriers are examined. Practical engineering codes are developed with the support of experimental findings gathered on dedicated facilities. It is shown that forced dispersion is more appropriate for the dilution of flammable gases, while the efficient dispersion of a toxic dense pollutant requires the use of inhibiting liquid curtains. The design rules of these two types of spray curtains exhibit important differences. Mechanical curtains require coarse upward sprays in combination with a small upstream fence. Physico-chemical absorbing barriers rely on fine sprays; therefore, additional studies of wind effect are needed. The concept of the dual diluting/inhibiting curtain is suggested for future investigation.

Probabilistic analysis of transient events by an event tree directly extracted from operability analysis

Abstract: Operability analysis (OA), developed as far asits ‘quantitative’ aspects, was employed to examine two critical features of a pressure-regulating installation (PRI) of a branched gas mains: ‘sudden blow-off’ and ‘continuous discharge of gas’ from the vent valve. An estimate was obtained of the probability of their occurrence by an event tree directly extracted from an OA, not handled in the classic manner, but treated as a tool for a sequential logical analysis. This full integration of qualitative and quantitative methods ensures the essential features for a probabilistic safety analysis, namely: systematism, completeness and congruence; numerical reliability and verification.

Quantification of escalation effects in offshore quantitative risk assessment

Abstract: Loss-of-containment events on offshore installations can lead to complex escalation effects, especially in congested modules, where the close proximity of processing equipment and control systems creates the potential for very many different escalation paths. Conventional risk analysis tools are not well suited to deriving and evaluating the large number of possibilities, relying on the risk analyst to devise and quantify a ‘representative’ set of scenarios. The danger is that the ‘representative’ scenarios lack realism, may be insufficiently detailed to support upgrade decisions, and tend towards ‘generic’ risk analysis. A novel computer simulation technique has therefore been developed which starts from a list of initiating loss-of-containment events and utilizes physical rule sets and consequence models to generate consistent and detailed accident scenarios from a basic platform description, comprising a collection of equipment items, structural elements and protective systems. The program has proved stable on application to a variety of offshore installations, including simple gas platforms, a typical deep sea oil and gas platform, and a semi-submersible production vessel. It eliminates the manual drafting and analysis of event trees, which is very laborious even when aided by interactive computer tools; and, because the risk model comprises factual information, rather than abstract concepts, it is much more accessible and amenable to scrutiny by discipline engineers, allowing operators to derive and maintain a genuinely ‘living’ quantitative risk assessment.

A summary of incident reporting in the process industry

Abstract: Accidents (resulting in injury, damage and other losses) can be prevented by acting on the information available from the reporting and analysis of incidents (including near-misses). The design and quality of a reporting system will determine its performance; however, with no standards and very little guidance from publicly available literature, how do companies handle incident reporting? This paper is a summary of the reporting systems of about a dozen companies from whom we have received copies of their report forms. It covers their contents and includes some details of the associated procedures. The main conclusion is that every form is different. The value of reports comes from how the information recorded is used. This depends on the quality of reporting and investigation, which will be determined by the safety management systems. No company covers everything mentioned in this paper. Some reports are more comprehensive than others, but we have yet to discover which factors give the best performance by actually preventing accidents. We suspect that more fundamental aspects of the company culture and general management will prove to have a very great effect that could result in a ‘good’ reporting system performing badly or vice versa.

Numerical simulation of heavy gas cloud dispersion within topographically complex terrain

Abstract: Numerical analysis has been performed for predicting the dispersion of heavy gas clouds released without initial momentum (i.e. dense gas puffs) within regions of complex topography. The three-dimensional non-steady differential equations governing transport are solved by means of the numerical finite volume method, using a collocated variable arrangement with a fully implicit integration over time. The turbulence effects on the flow properties are simulated by the two-equation k -ϵ turbulence model. Comparisons between calculated and measured data are presented, showing good agreement between them. The method is also used to predict a chlorine release within a fictitious industrial plant. The location of the release site, the atmospheric stability class and the wind direction are varied, in order to show the effect of topography on the cloud dispersion.

Combustion of some pesticides and evaluation of the environmental impact

Abstract: A method of combustion and sampling for small-scale laboratory experiments under defined conditions is described. The combustion products were analysed qualitatively and quantitatively, and their environmental impact and toxicology are reviewed. For sampling the organic combustion products, a three-step sampling system is necessary: a cold trap for cooling the combustion products, a filter to sample compounds which have condensed to particles, and an activated carbon tube to sample lower boiling compounds. The pesticides investigated in this study were 2-methyl-4-chlorophenoxyacetic acid, dimethoate and methyl parathion. During a real fire situation, the gaseous combustion products are the most dangerous to the fire-fighters and nearby residents. To protect the environment from disaster during a serious fire in a warehouse where pesticides are stored, the fire-fighters should perhaps not even try to extinguish the fire if it is widespread. Instead, they could protect the surroundings at the scene of the fire and could promote a fast hot fire to burn out all the organic material. This would raise the toxic gases higher into the atmosphere, where they would be diluted, thereby reducing the risk that residents living nearby would be exposed to very high, even lethal, concentrations of toxic gases. In this case, the environmental damage would also be minor, because the organic materials would not enter the soil and they would not be transported by fire-fighting water.

Evaporation rates of liquid hydrogen and liquid oxygen spilled onto the ground

Abstract: The evaporation rates of liquid hydrogen and oxygen spilled onto the ground surface were measured in laboratory tests. To simulate the ground, concrete, dry sand and wet sand layers were used in a vacuum-insulated cylindrical glass vessel. Based on the temperature variations within the layer and detailed observation through the side of the vessel, the heat transfer modes controlling the evaporation phenomenon were elucidated. When a wet sand layer was used, the liquid oxygen or hydrogen did not soak into the layer, because the frozen layer of water between the liquid and the sand layer acted as a barrier. When a concrete layer was used, the liquid vaporized above the layer. In these cases, the evaporation rates were inversely proportional to the square root of the time, except in the early stage just after the start of vaporization. This relationship could be predicted by a simple calculation of heat conduction within the layer. On the other hand, when a dry sand layer was used, liquid oxygen was observed to vaporize while constantly soaking into and rolling up the upper section of the layer. In this case, the evaporation rate was determined simply by the velocity of liquid penetration downward through the sand layer. When liquid hydrogen was used, the liquid did not soak into the dry sand layer, and the evaporation mechanism seemed to be the same as that for the wet sand layer, because the air contained within interparticle cavities in the dry sand layer solidifies at the boiling point of liquid hydrogen.

Relief vent sizing for a Grignard reaction

Abstract: The formation of a Grignard reagent is highly exothermic. In addition, the Grignard reagent reacts violently with many compounds, such as water, to generate heat and possibly result in the generation of volatile and flammable products. As a result, two potential upset scenarios during the manufacturing of a Grignard reagent were identified through a hazard and operability study. Bench-scale studies of these scenarios using the vent-sizing package (VSP) were employed to evaluate the consequences of these scenarios. Based on the VSP test results, two additional upset scenarios were identified and further evaluated by the VSP. Sizing of the emergency relief device for these upset scenarios was then conducted by utilizing pseudo-one-component tempered system analysis. The recommended emergency relief system was installed prior to start-up and was found to provide adequate protection after one of the identified scenarios (runaway reaction) occurred. The details and potential causes of this incident are also discussed.

Vent sizing for gas-generating runaway reactions

Abstract: Exothermic runaway reactions that generate non-condensible gas as the temperature increases, as is typical of decompositions for example, can reach extremely high rates of pressure rise necessitating emergency relief of the process vessel containing the reactant. Sizing of a relief device using presently recommended methods (e.g. DIERS) frequently leads to extremely large and expensive vents. This paper presents a methodology that leads to a simple but much improved method for vent sizing, fully allowing for two-phase release of the gas—liquid mixture. A number of examples are presented which lead to interesting conclusions about the influence of plant variables.

The modelling of pool vaporization

Abstract: This paper describes a mathematical model which calculates the time dependencies of the flow rate and composition of the vapour emerging from a pool. A large variety of accidental cases can be covered: continuous or instantaneous spills, on confined or unconfined ground, ideal or non-ideal liquid mixtures in boiling or evaporating conditions. The boiling, when present, is modelled through an equation system comprising the Rachford-Rice relation and the energy balance of the pool, which is assumed to be well-mixed. In the case of a volatile pool, interfacial mass rates are determined taking into account the Stefan flux, and the thermal resistance inside the liquid phase is also considered. In all situations, the energy balance includes the contribution of ground, sun and air. Known experimental data have been used to validate the model.

The influence of human factors science on safety in the offshore industry

Abstract: This paper aims to clarify applications of human factors science in the offshore process industry in terms of what can be done, why and how. Illustrative examples are given. An explanation is given of the meaning of ‘human factors’, and the use of human factors science before and after the Piper Alpha disaster is outlined. The influence of approaches developed after the nuclear Three Mile Island accident is discussed briefly. The need for human factors review of offshore installations in design, construction and operation is highlighted by providing quantitative information on underlying causes of loss-of-containment accidents. Human factors review should be part of the safety management system, which is described in terms of a set of control and monitoring loops; the control and monitoring requirements for optimizing human performance are outlined briefly. Using the concept design stage of an installation as an example, the use of human factors review in safety decision-making is exemplified by indicating which areas should be targeted for review and how this would influence safety. Key human factors review methods are identified and an example is given of one approach, a ‘walk-through’ of design and procedures. It is concluded that identification of causes of accidents has prompted the development of human factor applications, but that there is still room for much more comprehensive and long-term human factors programmes in the offshore industry, with considerable potential for risk reduction.

FIRE - A database on chemical warehouse fires

Abstract: In connection with an EC-funded research project concerning toxic effluents from chemical warehouse fires, a database containing detailed information on selected fire cases has been developed. This paper describes the database and gives examples of its application for identification of fire types, substances involved and fire products/smoke characteristics.

Methodological aspects of the application of adiabatic calorimetry for thermal safety investigation

Abstract: The method of mathematical simulation is used for analysis of some methodological problems related to the application of adiabatic calorimetry: correctness of the procedure of initial temperature determination, influence of thermal inertia on temperature distribution in a reacting system, features of data interpretation in the case of a complex reaction mechanism, etc. The effectiveness of the approach based on the use of mathematical simulation and appropriate software is illustrated by several examples.

Recent progress in the understanding of steam explosions

Abstract: If a hot liquid contacts a cold volatile liquid, a violent explosion may result. Such explosions are physical in character and involve a complex sequence of events. Initially, the liquids mix on a coarse scale and the system is relatively quiescent. However, if a suitable triggering event occurs, a violent shock-driven propagation can follow, which results in the generation of very high pressures and damage to surrounding structures. In this paper, we present a brief description of the various industries affected by such explosions and describe the underlying sequence of events which leads to these explosions. We then summarize the recent progress which has been made in this area and describe a new initiative which we believe will benefit all those with an interest in this phenomenon.

Auditing — a European perspective

Abstract: This paper describes the background research, development and application within the UK Health and Safety Executive (HSE) of an audit technique for assessing the effectiveness of safety management systems at onshore major hazard sites. The work is driven by HSE's use of quantified risk assessment for formulating advice to local planning authorities on the siting of major hazards and the development of land in their vicinity. The work aims to assess whether it is possible to use the auditors' assessment of the standards of safety management such that the effects of different perceived standards can be incorporated into risk assessment procedures. Further development and research are now the subject of a research project, funded by the Commission of the European Communities (CEC).

Dispersion in the presence of buildings

Abstract: Buildings can significantly affect atmospheric dispersion. There is an infinite number of possible configurations of release and obstacle arrangements, and no general theory exists to predict the dispersion of gas releases for all conditions. In this paper, two ‘classes’ of configurations, i.e. dispersion in the lee of buildings and dense gas dispersion over a fence, are discussed in detail and simple models are described. An example of the effects of multiple building arrangements is also presented.

Course and strength of accidental explosions on offshore installations

Abstract: To be able to limit the consequences of gas explosions, one has to predict gas explosion loads. Once these loads are known, structural measures to limit damage to the offshire rig can be taken. Prediction of the loads due to gas explosions is only possible on the basis of a good understanding of the processes governing this phenomenon. The aim of this paper is to describe the current understanding of gas explosion development in complex geometries. Important parameters of influence will be presented using experimental results. The principles of mitigation methods will be explained and methods for predicting gas explosion loads will be reviewed. Work performed at Christian Michelsen Research will be used as examples.

Case histories of incidents in heated bitumen storage tanks

Abstract: A study has been made of 73 case histories of incidents involving heated bitumen storage tanks. The majority of incidents were found to be due to operations such as charging and discharging of tanks, including overfilling which can lead to lagging fires and exposure of heater tubes. A number of incidents were the result of spontaneous ignition or unknown causes: it is proposed that regular cleaning of tanks would reduce the number of these incidents.

Structural response to accidental explosions and fires on offshore process installations

Abstract: The potential for a release of hydrocarbons and subsequent explosion and fire on offshore process facilities is relatively high This implies that explosion and fire events are normally defined as design accidental load situations This definition means that the facilities have to be designed to an acceptable level of safety to resist such loads in order to protect personnel, environment, facilities and production income A design for explosions and fires requires that protective measures have to be implemented These may, for example, lead to increased dimensions of the structural system, equipment protection, additional safety systems etc which increase the weight and costs of the facilities Extensive research and development programmes have been carried out during the last decade in order to establish reliable methods which can be used in design for explosion and fires This has led to improved simulation techniques and programs for the probability, loads and response to structures and equipment Furthermore, the extensive testing programmes which have been carried out have led to an improved physical understanding of the behaviour during explosion and fire exposure Such experience is of major importance to safety design work since it makes it easier to identify the real critical events, structures and equipment Another important issue for the safety design of offshore process facilities has been the introduction of safety acceptance criteria Such criteria may be expressed as upper acceptable risk levels for certain safety parameters such as personnel risk, damage to environment, damage to facilities, etc The use of safety acceptance criteria allows for more flexible design development compared to the use of traditional rigid design requirements

Wind-tunnel simulation. Boundary layer effects in dense gas dispersion experiments

Abstract: There is concern over the use of enhanced surface roughness values in the modelling of dense gas dispersion on process sites. We review the effect of roughness on dispersion and present the results of a set of dense gas dispersion wind tunnel experiments in which the roughness elements are large compared to the cloud depth. Measured concentrations and cloud widths are compared with the predictions of a dense gas model, HEGADAS, which was developed to predict dispersion over flat terrain. Model predictions are found to be in good agreement with measurement using appropriate values for the aerodynamic roughness length of up to about 0.5 m at full scale. Interactions between roughness elements and the flow field need to be taken into account when comparing model predictions for boundary layers with a greater roughness length.

Methodology for the analysis of risks during the construction and installation phases of an offshore platform

Abstract: This paper describes the development and application of a quantitative method for the assessment of risk of asset loss due to structural damage during the construction through to installation phases of an offshore concrete gravity-base structure. Risk criteria, embodying the ‘ALARP’ principle, are developed for different degrees of platform damage, expressed as asset loss. Initiating events leading to possibly significant failure modes are identified and quantified, mostly using fault trees. Much of the effort has gone into identifying and quantifying human error frequencies, including those for recovery failure, based on detailed scrutiny of the operations and established human factors methodologies. Outcome modelling for the initiating events uses an event tree approach, and the detail in the consequence modelling, e.g. for modelling of structural damage, is commensurate with the expected risk as identified in a ‘first pass’ analysis. Risk-reduction measures are targeted at dominant risk contributors, whose principal determinants can be identified from the fault trees. The application of the method to the analysis and mitigation of dropped-object risks during certain construction phases of the platform is described.

European model evaluation activity

Abstract: The Commission of the European Communities (CEC) launched an activity on evaluation of technical models used within the major industrial hazards area in November 1992. This paper describes the activities, including a presentation of the guidelines which have been prepared. The CEC commissioned a report by Dr R.E. Britter of Cambridge University on the evaluation of technical models used for major accident hazard installations, following which it was agreed to propose the establishment of a model evaluation group which would study the problem further. The objectives of a model evaluation group are twofold. The primary aim is to improve the culture in which models are developed and used and so ensure that models used in all aspects of major hazard assessment are up to date with technical developments and utilized by personnel well-versed in their applicability and functioning. Secondly, the results will be used to assist the CEC in establishing a balanced set of research priorities for its research programmes within major industrial hazards.

Pesticide warehouse fire consequences: experiences from a Danish risk analysis study

Abstract: A risk analysis of a pesticide warehouse has been carried out by Riso National Laboratory. The most important risks from the plant were emission of toxic substances released from fires involving large quantities of chemicals, and release of fire-fighting water contaminated with chemical substances. Two areas where more knowledge is required were identified: (i) assessment of environmental effects due to release of ecotoxic substances with the fire-fighting water; and (ii) assessment of the toxicological effects to humans of the fire effluents from a fire involving large amounts of chemical substances. With respect to the latter problem, important research areas are development of methodologies for assessment of the chemical nature and quantitites of the toxic substances that can be generated during the fire course, and also estimation of the effect to humans from inhalation of a complex mixture of fire products. In order to remedy some of the problems, an experimental research project has been initiated, sponsored by the CEC STEP programme. As an example, some results of the work are given in this paper: LC50 values are shown for the combustion atmospheres generated from selected pesticides by the DIN 53 436 furnace. The values are determined by use of an empirical N-gas model.