Predictive human error analysis

Predictive human error analysis can be performed manually or by means of a computer software package. Three types of analysis are possible within PHEA. 

Predictive human error analysis (PHEA) is the process via which specific errors associated with tasks or task steps are predicted. The process also considers how these predicted errors might be recovered before they have negative consequences. The inputs to the process are the task structure and plans, as defined by the task analysis, and the results of the PIF analysis. The basic procedure of the PHEA is as follows ... 

A. Perform Detailed Predictive Human Error Analysis (PHEA)... 

This section illustrates how the techniques described in Chapter 4 can be used to develop a procedure for the job of the top floor operator in the batch plant considered earlier. Two techniques are illustrated (i) a hierarchical task analysis (HTA) of the job, and (ii) a predictive human error analysis (PHEA) of the operations involved. HTA provides a description of how the job is actually done while PHEA identifies critical errors which can have an impact on the system in terms of safety or quality. The basic structure of the procedure is derived from the HTA which specifies in increasing detail the goals to be achieved. To emphasize critical task steps, various warnings and cautions can be issued based on the likely errors and recovery points generated by the PHEA. 

FIGURE 5.8 Results of Predictive Human Error Analysis. 219... 

Predictive Human Error Analysis in Permit To Work system in a petrochemical plant... 

ABSTRACT Permit To Work (FTW) is a means of safety system to coordinate different work activities. However, it may be susceptible for human error. The purpose of this study was identification and analysis of human errors in PTW system using Predictive Human Error Analysis (PHEA) technique. The most important identified errors were inadequate isolation of process equipments, inadequate labelling of equipment, and delay in starting the work after issue the work permit, improper gas testing, inadequate site preparation measures etc. Finally for preventing and recovering from the identified errors, site work permit form and its procedure was revised. 

Rasmussen, J. 1979. Notes on human error analysis and prediction. In G. Apostalakis and G. Volta (Eds.), Synthesis and Analysis Methods for Safety and Reliability Studies, Plenum, New York. 

In addition to their descriptive fimctions, TA techniques provide a wide variety of information about the task that can be useful for error prediction and prevention. To this extent, there is a considerable overlap between Task Analysis and Human Error Analysis (HEA) techniques described later in this chapter. HEA methods generally take the result of TA as their starting point and examine what aspects of the task can contribute to human error, hr the context of human error reduction in the CPI, a combination of TA and HEA methods will be the most suitable form of analysis. 

The application of human error analysis (HEA) techniques is to predict possible errors that may occur in a task. The next stage of error analysis is to identify error recovery possibilities implicit within the task, and to specify possible... 

The intention of this chapter has been to provide an overview of analytical methods for predicting and reducing human error in CPI tasks. The data collection methods and ergonomics checklists are useful in generating operational data about the characteristics of the task, the skills and experience required, and the interaction between the worker and the task. Task analysis methods organize these data into a coherent description or representation of the objectives and work methods required to carry out the task. This task description is subsequently utilized in human error analysis methods to examine the possible errors that can occur during a task. 

The technique for human error-rate prediction (THERP) [ Swain and Guttmann, 1980] is a widely applied human reliability method (Meister, 1984] used to predict human error rates (i.e., probabilities) and the consequences of human errors. The method relies on conducting a task analysis. Estimates of the likelihood of human errors and the likelihood that errors will be undetected are assigned to tasks from available human performance databases and expert judgments. The consequences of uncorrected errors are estimated from models of the system. An event tree is used to track and assign conditional probabilities of error throughout a sequence of activities. 

An evaluation method to determine the probability that a system-required human action, task, or job will be successfully completed within the required time period and that no extraneous human actions detrimental to system performance will be performed. It provides quantitative estimates of human error potential due to work environment, human-machine interfaces, and required operational tasks. Such an evaluation can identify weaknesses in operator interfaces with a system, quantitatively demonstrate improvements in human interfaces, improve system evaluations by including human elements, and demonstrate quantitative prediction of human behavior. See also ATHEANA (A Technique for Human Error Analysis) Human Error Analysis. 

In the process of risk and human reliability assessment, there are various methods to be used, such as Cognitive Reliability and Error Analysis Model (CREAM), A Technique for Human Error Analysis (ATHENA), and Technique for Human Error Rate Prediction (THERP). 

Process Hazards Analysis. Analysis of processes for unrecogni2ed or inadequately controUed ha2ards (see Hazard analysis and risk assessment) is required by OSHA (36). The principal methods of analysis, in an approximate ascending order of intensity, are what-if checklist failure modes and effects ha2ard and operabiHty (HAZOP) and fault-tree analysis. Other complementary methods include human error prediction and cost/benefit analysis. The HAZOP method is the most popular as of 1995 because it can be used to identify ha2ards, pinpoint their causes and consequences, and disclose the need for protective systems. Fault-tree analysis is the method to be used if a quantitative evaluation of operational safety is needed to justify the implementation of process improvements. 

These explanations do not exhaust the possibilities with regard to underlying causes, but they do illustrate an important point the analysis of human error purely in terms of its external form is not sufficient. If the underlying causes of errors are to be addressed and suitable remedial strategies developed, then a much more comprehensive approach is required. This is also necessary from the predictive perspective. It is only by classifying errors on the basis of underlying causes that specific types of error can be predicted as a function of the specific conditions under review. 

The use of a model of human error allows a systematic approach to be adopted to the prediction of human failures in CPI operations. Although there are difficulties associated with predicting the precise forms of mistakes, as opposed to slips, the cognitive approach provides a framework which can be used as part of a comprehensive qualitative assessment of failure modes. This can be used during design to eliminate potential error inducing conditions. It also has applications in the context of CPQRA methods, where a comprehensive qualitative analysis is an essential precursor of quantification. The links between these approaches and CPQRA will be discussed in Chapter 5. 

The various analytical methods for predicting and reducing human error can be assigned to four groups or sections. In order to make a start on any form of analysis or prediction of human error, it is obviously necessary to gather information. The first section therefore describes a number of techniques that can be applied to acquire data about what the worker does, or what happened in an accident. 

The other main application area for predictive error analysis is in chemical process quantitative risk assessment (CPQRA) as a means of identifying human errors with significant risk consequences. In most cases, the generation of error modes in CPQRA is a somewhat unsystematic process, since it only considers errors that involve the failure to perform some pre-specified function, usually in an emergency (e.g., responding to an alarm within a time interval). The fact that errors of commission can arise as a result of diagnostic failures, or that poor interface design or procedures can also induce errors is rarely considered as part of CPQRA. However, this may be due to the fact that HEA techniques are not widely known in the chemical industry. The application of error analysis in CPQRA will be discussed further in Chapter 5. 

In addition, the chapter will provide an overview of htunan reliability quantification techniques, and the relationship between these techniques and qualitative modeling. The chapter will also describe how human reliability is integrated into chemical process quantitative risk assessment (CPQRA). Both qualitative and quantitative techniques will be integrated within a framework called SPEAR (System for Predictive Error Analysis and Reduction). 

Qualitative human error prediction is the most important aspect of assessing and reducing the human contribution to risk. For this reason, it will be described in some detail in this section. The qualitative analysis performed in SPEAR involves the following techniques ... 

If the results of the qualitative analysis are to be used as a starting-point for quantification, they need to be represented in an appropriate form. The form of representation can be a fault tree, as shown in Figure 5.2, or an event tree (see Bellamy et al., 1986). The event tree has traditionally been used to model simple tasks at the level of individual task steps, for example in the THERP (Technique for Human Error Rate Prediction) method for human reliability... 

PROBLEM DEFINITION. This is achieved through plant visits and discussions with risk analysts. In the usual application of THERP, the scenarios of interest are defined by the hardware orientated risk analyst, who would specify critical tasks (such as performing emergency actions) in scenarios such as major fires or gas releases. Thus, the analysis is usually driven by the needs of the hardware assessment to consider specific human errors in predefined, potentially high-risk scenarios. This is in contrast to the qualitative error prediction methodology described in Section 5.5, where all interactions by the operator with critical systems are considered from the point of view of their risk potential. 

While the research described above suggested that the monkey was the species that yielded the most predictive correlations from which to predict human VD, Caldwell et al. have conducted a similar analysis and showed that VD data obtained in the rat yielded a predictive correlation [12]. In their approach, simply by multiplying the measured rat VDSS value by a factor of 188 yields a prediction for human (in units of volume that are not corrected for body weight) or by a factor of 0.67 when values are corrected for body weight. The approach yielded a mean-fold error of 1.85, which is a comparable level of error as other in vivo methods. 

A human error or reliability analysis (HRA) can be performed to identify points that may contribute to an accidental loss. Human errors may occur in all facets of a the hydrocarbon industry. They are generally related to the complexity of the equipment, human-equipment interfaces, hardware for emergency actions, and procedures for operations, testing and training. The probabilities of certain types of errors occurring are normally predicted as indicated in Table 29. Individual tasks can be analyzed to determine the probability of an error occurring. From these probabilities, consequences can be identified which detemline the risk of a particular error. 

Recently, Gasteiger et al. [59] reported several models to predict human oral bioavailability using Hou and Wang s data set. A set of ADRIANA.Code and Cerius2 descriptors were calculated, and MLR analysis was performed. The best linear model had r2 of 0.18 and RMSD of 31.15. When a set of subsets was cherry-picked so that each subset had either a common functional group or a similar pharmacological activity, the r2 values were improved and RMSD values dropped. But the performance of those models was still not satisfactory the standard errors were above 20.0 and r2 was lower than 0.6. 

---