Common forms of behaviorally experienced training include all of the following except

Structured Approach to Training Needs Analysis

Training and development are the means by which a basic level of individual and/or team competence is enabled. Safety critical work should include appropriate formal training in addition to on-the-job training and development. This may be in different forms, such as classroom-based training or simulation. The current thinking in learning and development uses the concept of the “70:20:10” model: 70% of development comes from performing job tasks and encountering problems that allow experimentation, practice, and honing of skills; 20% comes through social learning and coaching; and 10% of development is delivered through classroom and online learning. This model emphasizes the need for active, experiential learning rather than receipt of passive instruction. It is important to reflect on how training is increasingly moving to online solutions because it is sustainable, cheap, and easy to roll out. A large proportion of safety training is delivered online, and the remainder is generally presented in classroom settings. This suggests the need for the industry to refocus attention on delivering experience-based learning. This topic is readdressed later in this chapter and in Chapter 20.3 on Effective Supervision.

TNA is a structured approach, initially developed in the military setting (see Ministry of Defence, 2015), to assess training requirements and assess appropriate training methods to meet them. It is typically used to identify and support training needs created by the introduction of new or modified systems and equipment; it is particularly useful in the design stages prior to the introduction of the new system and fits within the CMS.

TNA is an iterative process and provides an audit trail for training related decisions. TNA is typically undertaken in five stages:

Scoping document;

Operational Task Analysis;

Training Gap Analysis;

Training Options Analysis;

Training Plan.

8.3.1.1 Training Needs Analysis

Training needs analysis includes the identification and evaluation of the training needs of the facility groups. Facility personnel’s job role and responsibilities are used as a basis for this analysis. An exercise to identify job changes and any problems related to skills should also be completed. Each job role and associated subtasks that make up the job role are analyzed in detail and a corresponding training base of support for skills is identified.

The training plan design and development can then be executed. The design of the training plan must consider the method to administer the training, factoring in some of the concepts presented in Figure 8.3, showing the Cone of Experience.

The training plan should also consider the way the training is intended to be rolled out. It must be done in a structured way that engages its targeted audience. This also distinguishes between “first time” training and refresher training. First time training will require a more in-depth focus with longer training course durations, for example. Refresher training is about future needs and the requirement to sustain the high-performing team’s knowledge and skills. There is also a requirement within the training program to bring on board or induct new employees. The training must be relevant and of high quality and involve a combination of classroom and experience-based training.

The key steps in development of the training program are illustrated in Figure 8.5.

Properly trained and performing facility integrity personnel are an integral requirement for ensuring facility equipment operates safely, effectively and efficiently. There are a number of elements associated with a holistic training plan.

14.6 Conclusion

We have described the instructional design process of developing and setting up a simulation session based on the ADDIE methodology. Each stage is indispensable: needs analysis, training design with the educational objectives, development with the educational tools and methods, setting up of the educational program with the actual implementation of a simulation session, and finally evaluation of the SBT activity. Any educational program in simulation must rely on a rigorous and systematic design process coordinated by a simulation educator, often with the support of a simulation operations specialist and a team of subject-matter experts who can contribute to the various educational aspects of an SBT activity.

14.17 Training: ACSNI Study Group Report

A review of training is given in the First Report: Training and Related Matters of the ACSNI (1990). Although this is a report to the nuclear industry, it is applicable in large part to the process industries also. The report starts from the point that human error is the major source of accidents and refers to the cases of Three Mile Island and Chernobyl.

The authors treat the topic under the following headings: (1) definition of training, (2) the safety culture, (3) initiation of training, (4) internal monitoring, (5) training needs analysis, (6) criteria for operator performance, (7) standards for training, (8) methods of training, (9) central vs site-based simulators, (10) individual vs team training, (11) training for stress, (12) training of management, instructors, etc., (13) certification and (14) privatization. They state:

It is misleading to think of training solely in terms of the transfer of items of verbal knowledge or technical skill from instructor to pupil. Although both knowledge and skill are necessary, they are not on their own sufficient to assure safety. They must be augmented by different qualities: habits of forethought and precaution that place minimization of risk first, and other goals such as short term performance or convenience, second.

The main purpose of training is to create a safety culture. The report rehearses some of the elements of this, which have already been described in Chapter 6. It is necessary that senior management give a lead and take certain specific measures. Line management should have operational responsibility inter alia for safety and training. There should be monitoring of the safety culture using objective measures and independent assessment and a policy of continuous improvement with specific targets and with feedback to the workforce. Training should aim to rehearse the individual's experience deliberately so as to reinforce compliance and self-monitoring and awareness and reporting of hazards. Senior management should ensure not only that the organization has a suitable formal training system, but that it is operating effectively in practice.

Training should be monitored. One method of monitoring is the use of objective measures such as the frequency of accidents, incidents, operational deviations, trips, etc. Another method is to sample regularly what people say about the safety culture and about their own attitudes. The monitoring should be done by an assessor who is independent both of the local line management and of the training specialists.

Task analysis should be used to establish training needs. The analysis should specify in concrete terms the information to be supplied, the alternative actions that can be taken, and the quality of calculation or judgement the task requires. It may reveal different needs such as knowledge about equipment or practice in control skills, and thus point to different types of training such as lectures or use of a simulator.

Evidence from a number of industries shows that, even after quite thorough training, an individual may have little idea of the hazards of the job. The training should aim to raise the awareness of these hazards.

The report states that performance on the job should be not only monitored but also measured. This implies that there should be criteria of operator performance. Closely related to this is standards for training. The report accepts that it may be unrealistic to seek totally objective measures. Where these are not practical, it suggests that the assessment take the form of two independent assessments by supervisory staff using carefully defined rating scales and that these staff themselves receive training to ensure that the criteria used are consistent.

With regard to methods of training, the report refers to the distinction between skill-based, rule-based and knowledge-based behaviour and between responses to signal and to signs. Control skills such as making adjustments to maintain constant a single variable have become relatively less important on automated plant, but there are a range of control room skills which still need to be learned. Skills used in normal operation may perhaps be adequately learned on the job, but this is not so for skills required in an emergency situation, for which simulator training is appropriate.

The report argues that for the control of nuclear plants, an average level of performance is not good enough. It follows that training may have to seek to change certain patterns of highly learned skilled-based behaviour.

The report places particular emphasis on the rule-based level of control. Here the operator responds to signs rather than signals and has to categorize the situation. Whilst rules may be learned to some degree in the classroom, practice in the use of rules to categorize the situation is most effectively given on a simulator.

The report gives high priority to the provision of simulators for each type of control room, since there are aspects of task that cannot readily be taught without them. It discusses the merits of local vs off-site simulators, but regards this question as less important that clear specification of the skills to be learned and the level of competence to be achieved.

With regard to individual vs team training, the authors suggest that the initial emphasis should be on individual training, with team training introduced later. Simulator training is treated as a form of team training.

Operator performance is affected by the level of stress experienced. It is not desirable that any operator of a nuclear plant should have to carry out the job under an abnormal degree of stress. This has implications for selection and for monitoring of individuals. The individual should be free from chronic stress symptoms and be able to tolerate acute stress. Given this, the training should give practice in performing the task in realistic conditions and with a raised level of stress so that the trainee learns to handle it. Factors that affect stress include the level of demand and the degree of control over the task that the person feels he has. The individual's confidence that he is in control may be built up by experience of successful handling of similar problems. This experience can be given using a simulator. The level of demands should be built up until it reaches that experienced in real operation.

Another aspect of stress is the danger inherent in a major incident. The reports acknowledge that this aspect is difficult to simulate in an exercise, but refers to the expectation of possible criticism. In a healthy safety culture criticism of the performance of an individual or team can be made and accepted constructively. The report urges that management at all levels should be periodically assessed for training need and that no-one should be ‘above it.’

The report devotes considerable attention to the question of certification, which is supported by some and opposed by others. In order to disentangle the argument, it is necessary to assign a clear meaning to the word ‘certification’. The report gives some eight possible definitions, together with comments. These may be summarized as follows: (1) specification of the content of training and clear responsibility for authorization of an individual to undertake certain actions; (2) separation between individuals who undertake (1) and line management; (3) a requirement that line management accept without further question separate assessment of the individual competence; (4) a requirement that line management authorize only persons from amongst those who have satisfied a separate assessment; (5) a rule that (3) or (4) apply only at the stage common to all reactors with different procedures thereafter; (6) a requirement that separate assessment should not mean simply testing by the training function but by another part of the organization; (7) a requirement that separate assessment should not be carried out by the licensee's organization at all but by a separate body and (8) a requirement that assessment should include observation of performance on the job and not merely verbal knowledge.

The report suggests that the first two interpretations would attract widespread consent. Opponents of certification may understand it in the third sense, but the authors doubt whether this view has proponents. A more serious argument turns on the fourth definition. Here opponents of certification argue that the trend is towards more involvement by line management in any matter which affects production and that there are dangers if it is not involved in training, whilst proponents emphasize that line management is liable to develop blind spots. The authors suggest that this points to the need for a ‘two key’ system, in which an operator needs to satisfy both an independent assessment and local line management. They argue that given the wide variety of reactor types in Britain the fifth definition would effectively rule out independent assessment. With regard to the sixth and seventh definitions, the report supports an assessment of the trainee performed by assessors who are independent both of the local line management and of the training function, but is unconvinced that the independent assessors need to come from outside the licensee. The authors support the type of assessment given in the eighth definition. With regard to privatization, the report makes the point that, in order to support the type of training envisaged, a certain minimum size of training establishment is required.

The framework for training developed in the report is therefore broadly on the following lines. The starting point is task analysis. On the basis of this a specification for training is formulated by line management and the training is carried out by the training function. It is monitored by assessors independent of both local line management and the training establishment. Assessment of the trainee contains a component of actual or simulated performance. The trainee first has to satisfy the independent assessors and is then authorized by line management. Assessment is a continuing process.

The report deals particularly with the training of process operators. However, it emphasizes that training should not be confined to process operators but should also cover instrument artificers, maintenance personnel, etc.

Under the guise of trainability testing, the report touches on the question of personnel selection. It states:

The Study Group believes it is important to find a valid way of deciding the relative success of different individuals, and that thereby it will become possible to devise satisfactory selection procedures. In other occupations it has frequently been found that systematic assessment of individuals can significantly improve trainability and subsequent performance. The potential of such methods should be explored further.

List of Abbreviations

ABS

American Bureau of Shipping

Human-Centered Design

Human Factors Engineering

Human Factors Integration

Hierarchical Task Analysis

Human Reliability Analysis

International Organization for Standardization

Norsk Sokkels Konkuranseposisjon

International Oil and Gas Producers

Target Audience Description

Training Needs Analysis

Tabular Task Analysis

User-Centered Design

Poor design of the human interface is a primary cause of human failure, whether it is the way information is presented, the layout of controls, the physical workspace, or aspects of the environmental design. Several examples were cited in Chapter 1, What is Human Factors?, demonstrating how poor design has contributed to major accidents. Specific design flaws for four major accidents are explained in Box 9.1.

Box 9.1

Specific Design Flaws Contributing to Major Accidents

Three Mile Island nuclear radiation disaster, 1979—Plant operators failed to diagnose the loss of coolant for several hours due to an ambiguous control room indicator on the user interface. It indicated the stuck valve as closed (the light only indicated the power status of the solenoid providing a false indication, on this occasion, of a closed valve).

BP Texas City refinery explosion, 2005—The board operator did not have a clear indication of fluid flows in and out of the isomerization tower or the actual level in the tower and consequently did not know about the dangerously high level.

Formosa chemical plant explosion, 2005—The plant operator erroneously tried to open the bottom valve on a reactor in process and then overrode the safety interlock releasing the reactor contents. The plant design contributed to both of these human failures.

Kegworth air crash, 1989—The pilots shut down a healthy engine due to no clear indication of which engine had failed. There had been changes in the design of the aircraft from the previous version of the Boeing 737, including the operation of the auto-throttle but also the design of a vibration meter which would have indicated that the wrong engine had been shut down.

Human performance is directly influenced by design. Something that is well designed achieves its purpose, is motivating and enjoyable to use and there is a lower risk of injury. Poor design on the other hand can lead to human failure, poor performance, frustration, and greater risk of injury. The more effort that is given to achieving an effective design, the less likely the need for “work-around” fixes during operations, the more reliable the human performance and the cheaper the overall cost of ownership. This case is argued in Chapter 10, Human Factors Integration Within Design/Engineering Programs.

A well-designed system requires an understanding of the end users, what they need to achieve, and effective design of the human interface to meet the end user needs. This includes their tasks, tools and equipment, workspaces, work environment, and organizational structures. All of these aspects of the system design need to enable the end users to execute the functions they need to perform.

Human Factors Engineering (HFE) is the application of human factors to the design of systems to enable the end users’ needs to be met. There are four iterative phases:

gain an understanding of what needs to be taken into account during the design using a toolkit of analysis techniques;

use human factors scientific knowledge embodied within standards and regulations in conjunction with the first phase to derive the design criteria;

design of the human interfaces within the system;

evaluate design to ensure it meets the needs and requirements of the end users.

Within this chapter, relevant terms and boundaries are defined prior to describing the complexity of human interactions and the approach for applying HFE to the design of new work systems. This is continued within Chapter 10, Human Factors Integration Within Design/Engineering Programs, to define how HFE needs to be integrated within engineering programs. The remaining chapters in this section are dedicated to specific HFE topics and areas of application. The section includes the following chapters:

Chapter 9—Overview of Human Factors Engineering

Chapter 10—Human Factors Integration Within Design/Engineering Programs

Chapter 11—Building and Control Room Design

Chapter 12—Workstation, work area, and Console Design

Chapter 13—Control System Interface Design

Chapter 14—Plant and Equipment Design

Chapter 15—Human Factors in Materials Handling

Chapter 16—Environmental Ergonomics

Chapter 17—Human Factors in the Design of Procedures

5.3 Lean TPM Implementation Roles

Throughout this book we have used the term ‘value’ as something we generate for customers. We have used it to describe processes that convert materials into saleable products, to describe processes that compress the time between receiving an order and fulfilling it – defining value from the perspective of the customer (Lean principle number one!). There is, however, another dimension to ‘value’ that is rarely discussed in the modern management literature and this is the value of organisational roles within the firm. An understanding of these ‘value roles’ is an important context that has a relationship with successful and sustainable process improvement.

The two generic work flows which add value as part of the top-down and bottom-up Lean TPM process are owned by the site lead team (top down) and the change team (bottom up). Collating the bottom-up audit results is used to assess top-down leadership progress through the Lean TPM master plan milestones as set in Figure 5.4.

6.6.2 Training system elements

A comprehensive training system should include full and accurate descriptions of the following elements.

Training and Commercial Food Handlers

In an attempt to improve food safety practices, legislation in a number of countries requires that all food handlers are supervised, instructed or trained in food hygiene matters commensurate with their work activities. This has in part resulted, in the UK, in an estimated 5 million food handlers receiving formal or certificated training over the past 10 years. This may only represent a fraction of all food handlers and there is evidence that businesses are still not complying fully with training requirements. Some countries are keen to encourage standardized food hygiene training. However, there is debate concerning the most appropriate type and nature of the training and whether introducing competence-based training is more, or less appropriate, than that delivered to meet the needs of various external awarding bodies. The latter are often private organizations and maybe international in nature. Endorsement of their courses may be seen as giving commercial advantage or preference to one organization or one country. However, some countries are less concerned with examinations and assessing the training and are more concerned with proof of attendance on training courses. Certificated formal training is preferred by auditors and inspectors because they can easily assess the level and content of the learning. Employers often prefer this approach because in some ways it is easier and allows them to display hygiene certificates and present a positive image. There is some evidence that employees prefer more specific internal, bespoke on-the-job training or supervision, believing it is more likely to lead to competence than more formal classroom-based training. Buddying can be used as part of on the job training/supervision, especially if based on a training-needs analysis.

Food service establishments are the main location for reported outbreaks of foodborne disease. It has also been implied that many outbreaks of foodborne disease result from faulty food-handling practices with one study suggesting that food handlers' malpractices contributed to 97% of foodborne illness in food service establishments and the home. This may be due to a lack of knowledge or nonimplementation of known food safety procedures. Effective training requires planned and systematic efforts to modify or develop knowledge, skills, and attitudes through learning experiences to achieve an effective performance in an activity or range of activities. Applied to food safety, this means people will receive, know, and understand food safety advice and be motivated to practice it. This is not just about sending people on a training course and hoping they will behave hygienically afterwards. To achieve behavioral change, people need to be motivated and be provided with the resources, including time, to act hygienically. The potential importance of training has been recognized over a number of years and recommended in many influential reports. Effective training offers, potentially, the best way to reduce cases of foodborne disease, yet it is often not well documented, managed, or supported, is taken for granted, and it is often seen by some enforcement officers and management as an end in itself rather than a means to an end.

Food handlers constitute a constantly changing and dynamic workforce, and some businesses may be reluctant to train staff fearing that they might leave and work for the competitors. Evidence of successful training can be seen in the form of hygiene certificates displayed in many food service establishments and small retailers, where they may be perceived to have a marketing potential and provide reassurance to customers. However, there is uncertainty regarding who should be targeted and the efficacy of current food hygiene training in changing handling practices and reducing foodborne disease. Many training initiatives have targeted the ordinary food handler rather than supervisors/managers, although there is evidence that businesses with better trained managers (who set standards and devise procedures) may sell food of better microbiological quality and are more likely to perform better in hygiene inspections. In a number of countries, training the supervisor or person in charge is seen as the most appropriate strategy. Training in food safety should involve all levels, not just food handlers and supervisors but also senior/top managers. They may need slightly different information with strategies designed to promote positive attitudes and culture as opposed to knowledge of specific practices. It is important for managers and supervisors to show food safety leadership. Statistics, although difficult to interpret, have not shown significant reductions in notified cases of foodborne disease as a result of all the training that has taken place. It is however possible that if the training had not been delivered, the number of cases could be even higher.

Catering or food service locations may be responsible for up to 70% of general outbreaks and it may be more difficult to ensure training is effective, i.e., behavioral change may be more difficult to achieve in this sector than in larger food manufacturers. Particular problems with ensuring training is effective in food service are summarized in Table 1.

Table 1. Characteristics of the food service industry that leads to training difficulties

It is now well recognized that the KAP approach to training is incomplete and greater use of other social cognition models is advocated. These argue that behavior is the outcome of an interaction between cognitive processes and environmental events, both of which can be affected by the overall training need and the views of interested parties. A model for effective transfer of training (Figure 2) and the role and influence of interested parties is presented. What people learn and whether they learn is influenced by training design and delivery, i.e., the course structure, information contained, physical learning environment, and quality of training. In addition, it is also influenced by their own internal motivation to learn and how this may be influenced by others, and their desire to comply with the wishes of others. The outcome from this is the individual's intention to behave hygienically, which can be again influenced by both the physical and social environment, the availability of the necessary facilities to practice hygiene, the operating procedures and the management systems they have to comply with, and whether there is a prevailing business culture to be hygienic. The latter are very important in determining whether the acquired learning is, or not, implemented. The sum of the individual food handler's practices gives rise to the collective operational food safety performance, which is an integration of the food safety management systems and food safety culture.

The likely effectiveness of training is a dynamic between the employer and the trainee. Employers should not automatically assume training will be implemented. The role of the food industry is not just about providing training, but it has very important functions in helping to create the desire for learning within their employees as well as providing the right culture/climate for this learning to be transferred into practice. This consists of identifying and maintaining behavioral food safety norms and standards, i.e., the level of hygiene expected. This is related to rewards for compliance and sanctions for noncompliance, as well as the provision of appropriate and adequate facilities to allow its implementation. Employers, and in turn employees, may display attitudinal ambivalence. Alternatively, employers may display a positive general attitude to hygiene and food safety, but they may be more negative toward the implementation of specific hygiene practices and be more concerned about making money. One of the greatest enemies of a positive food safety culture maybe an opposing economic culture where saving money is considered more important. For example, although good hygiene may be valued, managers may feel specific practices are too time consuming or inconvenient. In turn, food handlers themselves may perceive this desire to serve customers quickly or save money.

Another dilemma that may face industry concerns the type of training and who will deliver it. Options included work-based training, usually delivered in-house, or a more formal certificated training which can be delivered in-house or contracted out to one of the many training providers/consultants available. One of the concerns experienced by industry in the past, particularly with respect to externally accredited courses, has been their generic nature. This has often resulted in food handlers from manufacturing being trained with caterers or butchers and following the same general syllabus. The lack of specificity of this type of approach has been cited as one of the reasons why training can fail to change behavior. As a consequence, some countries now offer sector-specific syllabus and exams. In practice, the different sections may or may not be taught together. All trainers should provide relevant examples showing application of knowledge to specific sector work practices, although this may be difficult with large mixed classes. Some awarding bodies/training providers have attempted to provide bespoke courses for larger companies who can provide sufficient student numbers. Tables 2 and 3 provide a self-assessment checklist for employers to try to ensure maximum benefit from the training their employees receive.

Table 2. A model for training and behavioral change: factors influencing effective training

Table 3. Employer's self-assessment checklist for effective training

A frequently neglected component of training is the requirement for refresher/remedial or update training. Most training focuses on the critical need for food handlers when they start work. Yet for longer-term compliance especially in the light of new information, refresher/remedial training is essential.

From design to development

The systematic approach to design and development of instructional material places media selection well into phase 2 of the process. In theory, this phase of SAT—design, as shown in Figure (5)1—is concerned with formulating the learning strategy to be used in satisfying the enabling objectives established and deciding which medium is likely to be the most effective, efficient means of delivering the established strategies. The instructional material is constructed in small, easily handled units of study, which are placed in logical sequence representing learning blocks. Each of these blocks or modules of instruction deals with a particular learning objective. Depending on the type of learning involved (cognitive, skill or affective), a mode of teaching is available which provides effective instruction in this area, for example drill and practice, gaming, tutorial, simulation etc. The systematic approach dictates that only at this stage in the design phase is media selection dealt with.

Singh:

The inevitable conclusion … is that there is no single medium or delivery system which will solve all training problems. Effective and efficient training involves a mixture of media and methods, combined by the trainer into a blend which motivates the trainees and in which the strengths of each are complemented, while the weaknesses are cancelled out.

In practice, however, very few (if any) instructional systems are ever designed in this sequential manner. The medium to be used is usually a foregone conclusion and the designer is obliged to create material designed specifically for delivery by the existing, available means. This is not surprising since the existing medium proves to have two advantages over a proposed new one: it is established and it is understood. The emergence of TBT has not, even with convincing and influential cajoling, altered this situation.

Bhugra:

The rapid rise in the number of installed PCs has led some producers to ignore the primary criteria for media selection, namely the requirements set by the training objectives. Many CBT courses currently on the marketplace would be more effective if they had used text, video, tape/slide or other ‘conventional’ media. In reality a strategic decision to use CBT often precedes selection of the subject matter and is maintained, regardless of the subject matter. Occasionally other supporting media are included but this is becoming increasingly rare.

Rothwell (043):

If the training designer decides that tape/slide is the method most advantageous to the training requirement then, whether it is ‘readily available’ or not, it should be considered.

Singh:

The use of multimedia also affects the retention of the information that is transferred. The comparative figures for retention of information shown below are claimed to be typical of the results that have been found in many different surveys:

•

Textbooks = 30 per cent

•

Lectures = 40 per cent

•

Multimedia = 80 per cent/90 per cent.

The overriding considerations in preselection of media seem, then, to be tradition and understanding rather than effectiveness and efficiency. Those organisations which now use TBT methods, regardless of the reason, economic, political or commercial, have accepted technology-based delivery as the existing available means. For the most part they are preselecting TBT as the appropriate medium of delivery—in some cases, before the analysis phase has even begun. The disciples of the SAT prefer to explain this as special circumstances requiring a unique adaptation of the principles of the systematic approach to instructional design. This is, of course, true; the SAT at the height of its functionality is not rigidly suggestive of a final design. However, its disciples seem reluctant to acknowledge that what has emerged, in practical terms, is a SAT which, diagrammatically, looks like Figure (5)2.

Thus, media selection is entirely subjective and unrelated to informed judgements based on the analysis and design procedures. The analysis phase in this situation is purely a cosmetic adherence to principles. When viewed in its entirety, phase 1 of the SAT includes:

Training needs analysis

Task analysis

Target population analysis

Specifying objectives

Creating tests

Writing rule sets.

It becomes clear, if we are honest, that designers employ many ad hoc techniques, particularly in the early stages of requirement specification. This may be a result of the way of thinking encouraged by the SAT, in which the process of production is envisaged as a series of distinct steps, emphasising the difference between the phases, rather than showing them as different parts of the one whole. Despite warnings such as the following, the analysis phase still remains the most neglected.

Dean and Whitlock (005):

Intending authors of self-instructional materials will find the effort devoted to a thoroughgoing analysis well worthwhile; to skip over this stage will only cause problems later on. Authors who find themselves continually adjusting the wording and sequencing of their courses, adding new material and tinkering with exercises, are paying the price for neglecting or rushing the analysis process.

Shackle (050):

Much of the design process needs to be revised. In order to design for the non-expert user, the design process must not be computer centred nor application centred but user centred. The user and his tasks must be the starting point and continual focus for all design thinking.

Designers seem to be most at ease and therefore most productive when they have reached the stage in the design phase which calls for the creation of text comprising the actual learning material. Whether this is a video film script or a sequence of computer frames, designers often, mistakenly, believe that this is the essential part of the process and the easiest. Using a VDU monitor, confined basically to 24 lines horizontally and 40/80 characters vertically, as the communication device, naturally requires an awareness of ergonomics and an understanding of the means to effectively use facilities such as colour, sound and graphics. The issue of user friendliness applies to courseware as much as to any other software application.

Bhugra:

The limited display area is one of the most obvious constraints. Many developers choose to overcome the limitation of space by using successive screens to display text. This results in the commonly criticised ‘electronic page-turner’. A language specialist, however, recognising the constraint, can distil the message to fit the display limitations. The result is a sharper, ‘punchier’ instructional narrative. The editing skills of a ‘wordsmith’ are also invaluable in ensuring that grammar, spelling and punctuation are correct in the final drafts (an obvious requirement but often overlooked).

Bevan and Murray (015):

There are two essential requirements for successful interactive computer systems: they must be capable of reliably performing all the functions required by the user and they must be easy to use. No system will be successful if it does not have the required functionality and, unless they are very highly motivated, users will not tolerate difficult-to-use systems. ‘User-hostile’ systems flourish when systems are designed by computer professionals for computer professionals and when systems are intended for use by ‘captive’ users, employed specifically to operate the computer.

When trying to incorporate the facilities available, designers are easily drawn into misusing the computer's capabilities.

Steinberg (030):

Beginning authors tend to become enamored of CAI and try to see how creative they can be. They include every feature of the computer system even if it does not improve the lesson or actually detracts from it. For example, they design cute little characters to move across the screen, though the graphics neither motivate nor facilitate learning; such graphics may bore or slow down the student if they occur too frequently in the lesson. Authors may err by using many attention-getting techniques simultaneously, thus defeating the purpose of drawing attention to the critical feature on the display. For instance, they may use several colors, frames around words, text in several sizes and flashing, all at the same time.

In addition to the technical aspects of screen display techniques CBT creation is further complicated by instructional and interactive techniques. Unless the objective is to create an electronic book, issues such as the following cannot be ignored:

Questioning techniques

Answer judging

Feedback/response

Timing

Revision

Branching

Learner/computer control

Testing

Progress reporting.

Guest:

It has long been known in the psychology of learning that if a student is actively involved in the learning process, learning is more effective, retention is better and the learning process itself is described as more interesting by the learner. If the learner is passive, only 10 to 20 per cent of the material is absorbed, whatever the mode of training. However, retention can reach between 80 and 90 per cent when the student is actively participating. In correctly designed CBT, the student should be provided with every opportunity to actively participate in the learning process. A rule-of-thumb is that the student should be given practice items after every three screens of instructional material.

Rothwell (043):

For any training to be efficient and effective it is essential that a logical, planned approach to course design is made so as to provide the right training, to the right people, in order to give them the right skills. This is even more true of CBT than it is of the more conventional approaches. Some CBT courses do no more than reproduce on a screen that which could just as easily be read and learned from a book.

The implication is that extensive attention to detail in the analysis and design phases will guarantee that the actual scripting of the material will be essentially the conversion of the training principles already established into words, pictures and questions.

Before the advent of easy-to-use authoring tools it was commonly assumed that CBT material was, if not scripted by, certainly programmed onto the computer by, a computer expert (programmer). It was, therefore, necessary that the script was written in full, frame-by-frame, including instructions regarding display techniques and branching strategies to be used with each frame. Sometimes this process necessitated the use of specially designed screen layout forms, usually modelled on the original IBM IIS Worksheets, on which the designer would mimic what is to appear on the screen in detail. The worksheet method encouraged a rigid discipline and was meant to facilitate the development of the material into courseware. Since it is more likely, these days, that the same person will design and develop the course, very few bother with the creation of the entire course on paper before development. Indeed, some designers, encouraged by the authoring tool's sales literature, will design and develop, simultaneously, on-line.

How these two different procedures actually influence development time ratios depends surely on the designer's familiarity with CBT and with the authoring tool being used, and on the complexity of the course topic and material. The courseware development process will vary markedly according to circumstances and will therefore influence the total time taken from start to completion of the SAT. Having no access to case study information dealing with time ratios for the various phases in the SAT, we suggest that the following (assuming each phase is covered in its entirety by experienced CBT practitioners) indicates a reasonable breakdown:

Phase 1, Analysis—20 per cent

Phase 2, Design—20 per cent

Phase 3, Development—50 per cent

Phase 4, Implementation—10 per cent.

Raven:

The process can be streamlined by making the most productive use of subject matter expertise built up during the design phase. In other words, wherever possible the responsibility for CBT development should be assigned to someone who has already been involved with the design of a particular course. In the author's view, this is a powerful argument for the use of designer/developers rather than the more cash-effective (rather than cost-effective) approach of using designers to design and, subsequently, developers (lower cost) to develop.

Welham:

Simulation development for instructional and training purposes requires aspects of creativity and originality within what is in many cases a complex setting. Attempting to determine time and cost considerations for design, development and use of subject matter expertise for this type of project is very difficult. As in all cases, detailing procedures can reduce the scope for mistake. Production time forecasts for simulation development inevitably are less accurate than for CBT development (which in itself is not an easy activity to predict).

One final issue which seems to be changing, in practical terms, the process, and therefore the functions, of the development phase is the time and study given to the testing of the completed courseware before it is implemented as student-ready material.

Droar:

It may be possible to use some of the concepts from software engineering which allow final error rates (and hence quality) to be calculated from the number of errors found to date. However, as far as is known, the applicability of these concepts to courseware projects has yet to be demonstrated. A simpler system can be used whereby the number of errors found at each of the quality control steps (inspections and tests) is shown. However, great care needs to be exercised if this option is used as the figures can be easily misinterpreted. A further method involves comparing the number of actual inspections and tests performed to date against those planned, though it is arguable whether this is in fact a direct measure of quality.

Known variously as ‘internal validation’, ‘peer evaluation’, ‘pilot testing’ and ‘student try-outs’, this stage is theoretically placed as the last in the development phase or the first in the implementation phase. Its function is to allow a representative group of students to use the courseware as it was intended, in order to ensure that the product is ready to be distributed.

Heaford (008):

There is a temptation in all authors to get their newly developed course out into the world for all to see. A word of warning: make sure you fully validate your material in a CBT course before exposing it to an unsuspecting world. A bad experience in a new teaching method is likely to be doubly worse than a similar experience in a familiar medium. Mistrust in poorly delivered material will probably result in rejection of such methods for some considerable time afterwards.

Setting up these trials involves gathering the students, explaining the function of the test to them and to other ‘spectators’, arranging that the designer and/or developer is present and establishing guidelines for the analysis of the results. These requirements are influential in determining how often this particular stage is actually carried out. Other approaches to trials are more common—the review of completed material by the development team and the review of the completed material by the subject matter experts. There is vital information to be gained from these reviews, which may lead to improvements in the courseware before distribution—but what of the users and their role in the testing process?

Droar:

Acceptance test: at this stage the courseware should be largely error free, so the purpose of this last level of testing is to ascertain the acceptability of the final product to the ultimate user and/or the project sponsor. This type of testing tends to be much less formal than the previous two stages and may perhaps involve asking a sample group from the anticipated target audience to try the product and comment on its presentation, acceptability and so on. It should be noted that the distinction between acceptance testing and the final evaluation of the courseware can sometimes be blurred. However, it is important that some form of cognisance be taken of the reactions of the training's target group.

Steinberg (030):

The data [is] most likely to be reliable if you select students from the target population. If a lesson is intended for vocational students, test it with vocational students, not with professional students …. The first version of the lesson usually has many faults. Average or below average students are not likely to speak up, because they find it difficult to believe that deficiencies are the designer's fault if the lesson is hard or impossible to understand.

Incorporating pilot trials in the testing stage means that a further pilot trial is necessary if changes are made to the material as a result of a previous trial.

Dean and Whitlock (005):

In our ideal world you will make changes to the course as a result of the first validation run and then put another group of students through the same process. With a refinement process such as this, it is possible to end up with a course that you are confident meets the objectives set at the beginning, or such revised objectives as experience may have shown to be desirable.

5 OSH training and learning in SMEs – how is it done now, how should it be done?

Knowledge management and learning are closely interlinked as they are both concerned with the sharing and transfer of knowledge. The difference between the two is outlined by Schmidt (2005), who posits that knowledge management constitutes an organisational approach, while learning focuses on the employee in context. Schmidt’s view is that ‘what separates the world of e-learning and the world of knowledge management is their respective limited and isolated consideration of context’. The next section of this paper examines the literature on learning in SMEs.