Ever since the Three Mile Island accident, human-factor issues were largely associated with the design of the main control room. The possibilities of applying IIIE principles to improve human performance were limited to a great extent by the constraints of discrete, analogue instruments and controls. However, due to new capabilities offered by technologies like advanced sensors and automation systems, new NPP designs are now expected to introduce fundamental changes, not only in the design of the control room, but also in the role of the operator and the tools they use to monitor and control the plant. This could be regarded as a natural evolution for the industry, but it will require engineers and designers to rethink many tried and tested concepts and assumptions. For example, control centre structures need to be remodelled to make provision for new types of console and panel layouts, large screen displays, new communication media and even different crew structures. This will require a clear shift in the definition of the control room, its controls and instruments, its support structures and also the location of the control room in the plant.

Technically, it has become possible to control the plant from a remote location, but it will be a challenge to prove the reliability of such a scheme under all operational conditions. In addition to the changes in the physical and functional architecture of control rooms, we can also expect to see changes in the allocation of operational functions to humans and systems. The mere fact that future operators will deal with computer-based ‘soft controls’ and a multitude of high-resolution displays will already change their roles and mode of interaction with the plant. Where we today understand the operating crew as consisting of reactor operators, senior reactor operators and supervisors whose roles are largely determined by operating procedures, future operating crews may be regarded rather as part of the joint human-technology system, which in turn is part of the bigger socio-technical system of the plant. The reason for this lies in how the operator’s responsibilities and interaction with the plant will change. The shift will be more than just a role change due to increasing levels of automation, or an increasing supervisory role where operators’ primary function will be to monitor plant status and only to intervene if actual operation deviates from set points. Rather, there are now increasing possibilities for operators to perform ‘predictive control’ by examining past data, predicting future behaviour of processes by means of extrapolation and real-time simulation, and performing corrective actions before an event is likely to occur.

A further shift in the role of the operator will be an increase in the scope of responsibility and collaboration. For example, the scope of control and monitoring functions could increase from just operations, to include maintenance, production planning, and even design and optimisation.

All of these changes represent a paradigm shift for the nuclear industry, and it is almost entirely because of the advancement of automation and HSI technology. The changes have immediate implications for engineers who have to reconcile technological requirements with human abilities and limitations. There can be little doubt that automation is key to achieving cost-effective operations in future nuclear energy systems, but humans will continue to play as important a role in future systems as in today’s safe nuclear power plants. We can expect a different sort of HSI from that of today’s plants, but one in which the operator and crew are able still to intervene when necessary and otherwise oversee automation in many aspects of plant operation. This will require development of more ‘intelligent’ forms of automation and adaptive interface capabilities to facilitate near-autonomous operation as well as efficient human-system collaboration.

The following are some of the most important considerations that need to be included in power plant engineering and design strategies:

• The joint human-technology system must be defined in terms of the dynamic allocation of functions between the humans and the automation system.

• The human-technology system is not static and will require new rules and procedures for allowing a minimum number of operators to control multiple modules concurrently. Even for single module plants, it is possible that higher levels of automation will require fewer operators in the control room. However, regulators are unlikely to accept an unconventional staffing design without some kind of proof of concept. For new plants, this proof could be in the form of simulations or predictive computational models that provide reliable data on operator performance under various plant conditions (Persensky et al., 2005).

• Task support requirements — owing to the dynamic nature of the collaborative human-system relationship and the variable levels of complexity at different levels of automation, there will be variable requirements for task support. In principle, the lower the level of automation, the more the operator’s involvement in plant control and thus the more support is required, especially for non-routine tasks. HSIs that are designed to optimise human performance need to be concerned with fundamental collaborative functions such as coordination, adaptation and communicating shared awareness within the total socio-technical system. This goes beyond the present usage of computerised procedure systems, decision support, databases, data mining systems, and various devices to deliver this information to the user. The usability requirements for task support systems, especially those that use new HSIs, must include measures of the trust the operator places in the technology (Hugo, 2004).

These considerations suggest that HSIs can be examined from many different perspectives, but when we consider the challenges of emerging power plant designs, there are two main themes that seem to influence most considerations for future implementation:

• New HSI technologies offer innovative interaction modalities such as gesture control, augmented reality, remote control and telepresence. Designers need to provide, or obtain, sufficient evidence that these new concepts are conducive to usability and will support improved human performance.

• The applicability of advanced HSIs in the nuclear field is a particularly interesting question because the nuclear industry has been relatively stagnant for a long time. As a result, practices, standards, procedures and technologies have become so entrenched that utilities, vendors, regulators and other stakeholders have to make extraordinary efforts to justify and validate the use of new technologies. Even if those technologies had already shown proof of concept in other industries, the strict regulations and standards of the nuclear industry make implementation of any new technology an exceptional challenge.

The rest of the chapter will cover the most important aspects of HSI technology, starting with a description of the architecture or taxonomy of HSIs as they are typically deployed in ‘modern’ power plants. The following sections will also describe the range of technologies becoming available to designers, the technical capabilities they offer to support human performance, and the use and potential of a range of non-traditional HSIs, such as virtual and augmented reality systems, haptic devices and gesture controllers.