Microsoft Word - 6laurent.docx VOL. Guest E Copyrig ISBN 9 H Farz Centre naghd Huma and c optim perso The IS faciliti practic with th with t activit they w standa turns holisti This g consid should and th often aspec condit Strong good wheth guida engine decisi This p contin needs resea 1. Hu The d the sy transp attent comm CHE . 36, 2014 Editors: Valerio Co ght © 2014, AIDIC 978-88-95608-27- Human F Need fo zad Naghd e for Innovative dalf@tcd.ie an Factors En construction o ise the huma nal or process SO standard es projects ap ce this means he end users the scope of ties, including will use to com ards need to generates a n c approach to guidance shou derations to o d help avoid m he possibility a need to de cts can result tions at work. g operational design require her the curre nce for a fram eers, ergono ons to be mad paper will star nued by discus s of case stud rch. uman Factor demand of ind ystem from pr port systems, tion towards missioning and EMICAL EN ozzani, Eddy de Ra C Servizi S.r.l., 3; ISSN 2283-92 Factors E or More ali*, M. Ch Human System gineering (HF f socio-techni n contribution s safety or env ISO 9241-210 pply the princ s ensuring, as to identify wh the design p physical and mplete those ta be generic en need for more o guide Desig uld help stake ptimise and g more costly in of undesired emonstrate tha t in a major However, this performance es a detailed k ntly available mework to be p mists, project de in preparat rt by observing ssing what is dies taken fro rs Engineer ustry for safe rimary actor t manufacturing human-mach d sporadic su NGINEERI ademaeker 16 Enginee Structur hiara Leva, ms, Trinity Colleg FE) focuses o cal systems. n to production vironmental pe 0 (2010), Ergo iples of Huma s a minimum, hether there a project that ju d cognitive erg asks, and the nough so as to specific guida gners, Operat eholders ident uide some of ntervention lat events relate at this small i reduction in s is often only can only start knowledge of e standards a practically use t manageme tion and execu g and analysi currently ava m the industr ing impact i and efficient to supervisor g production li ine interactio upervision (Le ING TRAN ering at D red Guid Nora Balf ge Dublin, Dubli n the applicat The objective n and minimis erformance (O onomics of hu an Factors En that every ne are any “hotsp stify further H gonomic asse environment o avoid being ance on differ tors, Risk Ass tify and recog the solutions ter on in the li d to miss-con nitial investme the operatio demonstrated t with good de how humans and guideline ed by process nt and opera ution of projec ng the proble ilable and per ry. Addressing in Design: r operations ha of an automa nes and comp n is always eva et al., 20 NSACTION Design S delines a fe, Samuel in 2, Ireland tion of human e is to ensure se potential fo OGP, 2011). uman-system gineering (HF ew facilities pr pots” (risks, iss HFE activities essments of th in which they tailored to an rent processes sessors and P gnise the valu devised in the ifecycle of the nceived Huma ent in enginee onal life-cycle d retrospective esign and an u interact within es of the pro engineers, dis ational/mainte cts? m and presen rforming a sm g those gaps elevance of as increasingly ated process. puterized syst required eve 12). Additiona NS o Stage: Is and Stan Cromie n factors know e systems are or design-indu interaction, re FE) during ear roject is scree sues or oppor s. Further sta he operator ta will be undert ny specific des s and activities Project Planne e of Human F e early stages e product or p an Machine In ering to consi costs and i ely after minor understanding n the work sys ocess industry scipline engin nance line m nting example mall gap analys will be within f the problem y shifted the r This is partic ems; however en if it is jus ally, when the A publica The Italian Asso of Chemical Engin www.aidi s There ndards? wledge to the e designed so uced risks to h equires that a rly design stag ened in collabo rtunities) asso ndards detail asks, the equi taken. Howev sign process; s supporting a ers at design Factors Engin s. Such an app plant being de nteractions. Th ider Human F mprovement r or major acc g of what cons tem. The prob y provide su eers, human f management es. The study sis against co the scope of m role of the hum cularly true fo r a certain deg st for mainten e complexity ation of ociation neering ic.it/cet a ? design o as to health, all new ges. In oration ociated these ipment ver, the this in a more stage. neering proach signed here is Factors in the idents. stitutes blem is fficient factors during will be oncrete f future man in r rapid gree of nance, of the DOI: 10.3303/CET1436097 Please cite this article as: Naghdali F., Leva M.C., Balfe N., Cromie S., 2014, Human factors engineering at design stage: is there a need for more structured guidelines and standards?, Chemical Engineering Transactions, 36, 577-582 DOI: 10.3303/CET1436097 577 system increases, the reluctance of the designers to substitute the operator with automated functions will increase as well. This is because the ability of the human to control the system in unforeseen circumstances can help the system to keep functioning normally (Hale et al., 2007). Computers do not have this ability and therefore cannot be considered as the only available source of control. Therefore it is preferable to have the human as a final authority working with the computerized system. Hence, recently in most of the industries the human operator’s task to operate and control the system is considered crucial (Nazir et al., 2013). On the other hand there are accidents in which the human operator is overloaded with information and alarms that will affect their performance (ATSB, 2013). In accident investigations, design inadequacies are often mentioned as a major contributing factor (Hale et al., 2007) and human error is almost always described as a major cause of accidents (OGP, 2010). Considering the change of the human role to a supervisory role, the human-machine interface (HMI) is becoming an increasing risk for industries, one which better design can play a significant role in managing. Therefore, it is required to have the human machine interface carefully designed to meet the operator requirements and provide information and procedural guidance for the operator to improve the functionality of the system (Øwre, 2001). In fact the design of machinery and equipment and safety can no longer be considered as two separate tasks (Bernard and Hasan, 2002) and human factors engineering and ergonomic studies can help companies identify practical solutions for issues regarding the HMI (EU-OSHA, 2006). An example of inadequate human-machine interface design comes from the Three Mile Island incident which revealed the impact of HMI and procedure design on human reliability (Kim, 2001). Key indications were not visible to the operators leading them to follow a procedure which escalated the situation. This accident raised the question about the efficacy of the human factors supports in standards for designers and maturity of the practices (Boy and Schmitt, 2013). International standards are available to guide the application of HFE in the design phase (discussed in the following section) but these are by necessity general and might not be sufficient to support the HFE assessments of the design. This is illustrated by the fact that high reliability organizations in safety critical areas, such as the aviation or nuclear industries, have often developed their own internal standards to provide more specific guidance on HFE assessment and safety by design issues. There are several attempts in the field of design for improved approaches such as Human-Centred Design (Maguire, 2001), intelligent human-machine interface design (Tendjaoui et al., 1991), user needs analysis (Lindgaard et al., 2006), Safety by Design (Kletz, 1996) and Human Factors Integration (Widdowson and Carr, 2002). The intention of these approaches and methodologies is to prevent accidents and eliminate the source of the hazard as well as improving efficiency and well-being. There is also the possibility to provide more support in the design phase regarding human factors by learning from accidents and incidents. Different industries can learn from each other and historical data can contribute to increasing safety but the learning mechanism needs to have a good understanding of the culture, constraints, objectives and the design procedure of the target industry (Drogoul et al., 2007). Although learning from accidents and analysing them is very important for improving knowledge, on the other hand the learning procedure and the loop to give feedback to the engineering strategy is slower than technology developments, thus the learned lessons may lose their value and become inefficient (Rasmussen, 1997). In addition, new technologies are already introducing new challenges for the safety by design techniques in that there may not be any historical data available for them. Similarly market and financial issues are putting more pressure on the designers, thus the opportunity for thorough safety through design studies is decreasing (Leveson, 2011). In the past the development of a new technology was much slower than that in the present and it could allow enough time for the hazards to emerge (Leveson, 2011). 2. Existing standards and industrial practices for HFE in Design To support the challenging task of the design team there are number of standards able to provide some guidance on the minimum requirements in terms of human centred design: for example, ISO 6385 – Ergonomic Principles in the Design of Work Systems (2004) outlines how technological, economic, organisational, and human factors can affect the work behaviour and well-being of people within a work system. The general principle underlying the standard is that interactions between people and the components of the work system (e.g. tasks, equipment, workspace and environment) should be considered during the design stages. Each design stage is described and appropriate ergonomic principles and methods for each stage are listed. ISO 11064 - Ergonomic Design of Control Centres (2006) provides nine principles for the ergonomic design of control centres and guidance on specific aspects of control room design, including layout, workstation design, controls and displays, and environmental requirements. ISO 12100 – Safety of Machinery (2010) suggests a five step methodology to perform risk assessment at 578 design stage and the overall strategy requires designers to take into account the safety of machinery for their whole life cycle, considering usability, maintainability and cost efficiency. EEMUA 191 (1999) is an industrial standard developed by the Engineering Equipment and Materials Users’ Association to support the design of alarm systems taking into account the requirements of the human operator receiving and responding to those alarms, while EEMUA 201 (2010) is focussed on the design of HMIs and gives guidance on areas such as display hierarchies, the design of the screen format, and the attributes of the environment which may affect the use of the HMI. These standards define the minimum requirements and it is the decision of the designers on how to optimize and utilize the systems to increase the satisfactory level of the users. This systematic approach is fairly generic and does not provide technical support for the designers. While it recommends foreseeing the design uses nonetheless there is no discussion regarding the methodology to conduct this verification. Increasingly rapid prototyping and participatory approaches are proposed as methods to evaluate the design. These approaches have been commonly used for products that will be produced in large numbers (Sinclair, 2005), although it has traditionally been more costly and time consuming to apply this approach to the design of a control room, limiting the ability to apply these methods in this context. However a possible substitution for prototyping can be provided by the use of 3D models of the buildings, structures, or control room. Reviews of these models can be undertaken with the involvement of the operators. The 3D model is a more natural representation that does not require decoding of 2D technical drawings and thus facilitates the operator in identifying potential issues regarding the proposed design. This approach can be considered as an example of human centred participatory design, able to support a better understanding of the user’s needs and a more solid starting point for the designers to deliver a safer design. Such participatory reviews of designs do not negate the need for guidance for designers at an earlier stage as they should be facilitated as early as possible in optimising their design for human operation. The above-mentioned standards can be used in combination with 3D participatory review, however the process has not been detailed or suggested clearly in any of the before mentioned standards. So while on the one hand the ISO 9241-210 (2010), Ergonomics of Human- System Interaction, requires participatory human centred approaches it does not provide technical details on what specific aspects should be considered and how to concretely carry out such a process, the link with the more specific standards such as ISO11064 for the Ergonomic Design of Control Centres and or the ISO 12100 (2010) on Safety of Machinery is not structured or suggested in any clear way and as a result companies must introduce internal standards to tackle the problem. Table 1: Summary of the HFE issues in various areas of System Design HFE Area of Design Related existing standards / best practices Possible issues/ gaps Design of physical built environments ISO 6385 (2004) Ergonomic principles in the design of work systems The standards do not provide any practical guidance on how to actually review the built environment at the design stage involving users (such as 3D reviews) Design of machinery / electrical systems ISO 12100 (2010) Safety of machinery / EEMUA 178 (1994) A design guide for the Electrical Safety of Instrument Control Panels The standards are seldom applied in the industry and they do not specify to what machinery they should apply Design of control rooms, HMI for information systems EEMUA 201 (2010) / ISO 9241-210 / ISO 11064 (2006) Ergonomic design of control centres How to review the mimics of control centres is not specified and the use of task analysis is not clearly suggested Design of information systems and alarms EEMUA 191 (1999) / ISO 11064 (2006) As above Workload assessment for design ISO 11075-3 (2004) Ergonomic principles related to mental workload Not really applied in the industry Design of manuals and procedures ISO 12100 (2010) / ISO 18152 (2010) Ergonomics of human- system interaction – Specification for the process assessment of human-system issues The standards specify how to assess processes but not how to translate them in to good instructions and procedures Risk assessment at design stage ISO 31010 (2009) Risk management – Risk assessment techniques Little guidance on what standards are available for human reliability analysis 579 3. Examples of issues in a concrete case study In order to emphasise the importance of HFE in this chapter a few concrete examples of the application of Human Factors during the design of a gas plant will be presented. The selected case studies come from oil and gas design projects and have been selected because the human role has a significant importance in the process industry. The operators and the maintenance crew are the two main groups of users that interact with the system. As mentioned in the previous section, the insufficient detail in the publically available standards has led some companies, including those involved in the case studies, to develop internal standards to pursue safety during design. The aim of industry standards is to provide guidelines and support for designers and define general rules and decision criteria during the design review. In this case, the system was reviewed in a set of workshops and each HFE workshop aimed to detect the relevant issues regarding the human and system interaction. This paper will present the reviews relating to cognitive ergonomics of human-machine interaction (HMI), the 3D model, and alarm management. The first example is the evaluation of the human machine interface for a new control room within an oil and gas facility. The control room was designed to enable the operators to control the processes and functions of the facility and perform the necessary actions when required. The review of the control room was undertaken in accordance with ISO 11064 (2006) International Standard as well as the company internal standards. The review was made in a workshop with a team of experts using a checklist that considers the control panel in order to evaluate the design in terms of the operator’s ability to control the system through the designed control panel effectively and efficiently. During the cognitive review the HMI was reviewed to check that it was clear and understandable for the operators. This covered aspects such as overall system authority, information requirements, conformity with operator mental models, information coding, system feedback, and dialogue structure. However the review cannot be completed without running a test through some of the main tasks the operator has to perform on them. The need to supplement the guidelines from international standards reveals that although the ISO 11064 (2006) is a good starting point to set the design strategy it may not be adequate in terms of supporting the entire evaluation in the review sessions. The available standards do not have any recommendation for structure of the system to suggest recommendations, thus the company designed a recording system themselves. This system not only records the recommendations, suggestions, or action items but also describes the comprehensiveness of the feedback of review team and justifies the causes of the change. The recording and reporting mechanism for communicating with designers regarding the recommendations also relies on making points on a printed picture of the display as the recording system cannot provide all the necessary information for the designers in case of the display screen. The next case was alarm management within the process plant control system. In order to support the design of the alarm system, a private industry standard has been developed. This standard regarding alarm management is derived from EEMUA 191 (1999) and the structure of the review team is based on contribution of designers, HFE experts and operators. The alarm criteria are defined in the internal standard and for each specific alarm the designers have defined the specific parameters. In general, the priority of the alarm is based upon whether the operator has to act, and how much time they have to act before escalation of the event. The review team analysed each alarm that can be generated to make sure that the alarm is necessary (i.e. that the operator must act upon it) and that the operator has enough time to perform any necessary actions before the event is escalated, e.g. the designed instrumented protective function (IPF) or instrumented protective system (IPS) began to start. To make such analysis, the internal standard of the company provided a flowchart to structure the analysis and also investigate the necessity of the alarm. During the workshop, it was clear that the system designers and engineers had not considered the HFE principles of alarm management when designing the system, and in fact it is necessary to begin such workshops with a short presentation of the principles in order to ensure that all participants are aware and understand them. At times, the engineering standards they relied upon appeared to explicitly contradict the principles of alarm management creating situations which required complex solutions. The clarity of the HFE principles is critical in these situations to ensure that such situations can be satisfactorily resolved. However the alarms are always considered individually while in reality some scenarios may generate multiple alarms. The last example refers to the design review of built environment derived from the P&IDs. The company practices a participatory design approach because it was considered the best practice to accomplish a better design. Also the company adopted the use of 3D model to review the design as the use of P&IDs alone may result in missing some issues (e.g. low point, accessibility, maintenance issues etc.). A 3D graphical representation of the system can be employed to conduct the HFE review session. In the case study chosen the review took place in a brainstorming session among the designers, human factors experts and operators. This enabled the review team to find issues that may be discovered only when 580 considered in combination with actual tasks (e.g. the accessibility and visibility of key components, the safe positioning of equipment and vents, and the clear labelling of equipment). Those findings had a clear impact on improving the overall safety of the plant but could not have been easily identified without the use of a 3D model. Although in ISO 9241-210 (2010) it is mentioned that the users are to be involved nevertheless the end user’s involvement in the design procedure or any other alternative approach to take into account the user’s requirements was not delivered to support the designers. 4. Possible areas of improvement in HFE at the design stage Today’s sociotechnical structure is very complex and interrelated and requires a new approach to safety. Companies are facing difficulties in achieving a balance between safety, time and budget and a new oversight regulation or standard methodology could help them to ensure the safety more effectively (Leveson, 2011). Additionally, designing a complex system is a difficult task and designers need support. The standards discussed in this paper provide the basis for that support by giving guidelines and suggested approaches, but standards are necessarily broad in order to deliver their support to the widest possible range of end users. The need for more detailed guidance to support the design process is evidenced by the additional material used internally by companies to tailor the international standards to their operation. Although much of the available guidance is available to engineers and design teams (often it is specifically targeted at these groups), the ongoing need for detailed review sessions reveals that these groups are not fully assimilating HFE information. Integration of HFE principles within broader engineering and design standards may be one way to achieve this. Too often, only human factors specialists are aware of the existence of HFE standards and the principles contained within them. This means that the design reviews may be the first point when human factors principles are considered in the design, when in reality it should be a check point to ensure that they have been applied correctly. It is also important to ensure that the HFE standards are aligned with the relevant engineering standards, to ensure that designers are not receiving conflicting guidance from the two sets of standards. In order to best achieve this, engineers and designers should be provided with basic training in HFE to ensure that they understand the basic principles and are capable of correctly interpreting the information contained within the standards and applying it to their designs. In addition to the use of standards, designers can benefit from a clearer understanding of how their designs have performed in operations. As discussed, importance was placed on capturing the decisions made during the review meetings in sufficient detail to provide meaningful feedback for the design team. However, although the design review can decrease the number of hazardous situations or mitigate their effects, it cannot detect all the risks associated with a new design and these risks may be replicated in the future design projects conducted by the team despite being known in the operations field. Most design teams conclude their work following the implementation of the design or at best after the closure of the snagging list during the initial operations. They therefore do not have the opportunity to learn lessons from the operation of their design, and understand how their design is influencing operations. A new HFE method that can close this design loop and provide operations feedback to future design teams could provide very valuable design input, increasing overall safety and efficiency. In this paper the authors have tried to point out the gaps to be addressed by HFE at design. In future the objective of the study is to collect more data regarding the available methods and practices in industries to study them, justify the added value of implication of the methodology by industries and covering their weaknesses to support designers and increase the reliability of the systems. References ATSB, 2013, ATSB Report: In-flight uncontained engine failure of Airbus A380-842, VH-OQA. Australian Transport Safety Bureau, Canberra, Australia. Bernard A., Hasan R., 2002, Working situation model for safety integration during design phase, CIRP Annals - Manufacturing Technology, 51, 2, 119–122. Boy G. A., Schmitt K. A., 2013, Design for safety: A cognitive engineering approach to the control and management of nuclear power plants, Annals of Nuclear Energy, 52, 125–136. Drogoul F., Kinnersly S., Roelen A., Kirwan B., 2007, Safety in design – Can one industry learn from another?, Safety Science, 45,1-2, 129–153. EEMUA 178, 1994, A Design Guide for the Electrical Safety of Instrument Control Panels, EEMUA, London, UK. EEMUA 191, 1999, Alarm Systems: A guide to design, management and procurement, EEMUA, London, UK. 581 EEMUA 201, 2010, Process plant control desks utilising Human-Computer Interfaces, EEMUA, London, UK. EU-OSHA, 2006, European Risk Observatory Literature Review - The human machine interface as an emerging risk, European Agency for Safety and Health at Work, Bilbao, Spain. Hale A., Kirwan B., Kjellén U., 2007, Safe by design: where are we now?, Safety Science, 45,1-2, 305– 327. ISO 10075-3, 2004, Ergonomic principles related to mental workload - Part 3: Principles and requirements concerning methods for measuring and assessing mental workload, The International Standards Organisation, Geneva, Switzerland. ISO 11064, 2006, Ergonomic design of control centres, The International Standards Organisation, Geneva, Switzerland. ISO 11075-3, 2004, Ergonomic Principles Related to Mental Workload, The International Standards Organisation, Geneva, Switzerland. ISO 12100, 2010, Safety of machinery - General principles for design - Risk assessment and risk reductio, The International Standards Organisation, Geneva, Switzerland. ISO 18152, 2010, Ergonomics of human-system interaction - Specification for the process assessment of human-system issues, The International Standards Organisation, Geneva, Switzerland. ISO 31010, 2009, Risk management - Risk assessment techniques, The International Standards Organisation, Geneva, Switzerland. ISO 6385, 2004, Ergonomic principles in the design of work systems, The International Standards Organisation, Geneva, Switzerland. ISO 9241-210, 2010, Human-centred design for interactive systems, The International Standards Organisation, Geneva, Switzerland. Kim I. S., 2001, Human reliability analysis in the man–machine interface design review, Annals of Nuclear Energy, 28,11, 1069–1081. Kletz T., 1996, Inherently safer design: The growth of an idea, Process Safety Progress, 15, 5–8. Leva M., Pirani R., De Michela M., Clancy, P., 2012, Human Factors Issues and the Risk of High Voltage Equipment: Are Standards Sufficient to Ensure Safety by Design?, Chemical Engineering Transactions, 26, 273–278. Leveson N. G., 2011, Engineering a Safer world, MIT Press, Massachusetts, USA. Lindgaard G., Dillon R., Trbovich P., White R., Fernandes G., Lundahl S., Pinnamaneni A., 2006, User Needs Analysis and requirements engineering: Theory and practice, Interacting with Computers, 18, 1, 47–70. Maguire M., 2001, Methods to support human-centred design, International Journal of Human-Computer Studies, 55, 4, 587–634. Nazir S., Colombo S., Manca D., 2013, Testing and analyzing different training methods for industrial operators: an experimental approach, 23rd European Symposium on Computer Aided Process Engineering, 667–672. OGP, 2010, Risk Assessment Data Directory, International Association of Oil and Gas Producers, Report No. 434-5, International Association of Oil and Gas Producers, London, UK. OGP, 2011, Human factors engineering in projects, International Association of Oil and Gas Producers, Report No. 454, International Association of Oil and Gas Producers, London, UK. Øwre F., 2001, Role of the man–machine interface in accident management strategies, Nuclear Engineering and Design, 209, 1-3, 201–210. Rasmussen J., 1997, Risk management in a dynamic society: a modelling problem, Safety Science, 27, 2- 3, 183–213. Sinclair M. A., 2005, Participative Assessment, In Wilson J. R., Corlett N., Evaluation of Human Work, 3rd ed., 83–111, Taylor and Francis Group, Boca Raton, USA. Tendjaoui M., Kolski C., Millot P., 1991, An approach towards the design of intelligent man-machine interfaces used in process control, International Journal of Industrial Ergonomics, 8, 4, 345–361. Widdowson A., Carr D., 2002, Human factors integration : Implementation in the onshore and offshore industries, Health and Safety Executive, Camberley, UK. 582