The introduction of new technologies into heavily regulated industries such as aerospace is often a very complex, time-consuming, and expensive challenge that requires significant levels of research and development in order to ensure a successful technology substitution. This challenge is exacerbated when new technology options represent a fundamental shift away from well-established principles, as the risk and uncertainties involved increase significantly. This is currently the case in the anticipated transition from conventional turbojet aircraft architectures to all new electric configurations, and equally for the adoption of technologies enabling mass manufacturing and customisation processes in aerospace production lines. At the same time, the opportunities associated with these disruptive innovations may be sufficient to warrant decision-makers adopting new technological paradigms. In some cases, new technological frameworks arise even while existing technologies are still undergoing further developments, and have not yet reached the peak of their performance. This further complicates the decision for enterprises, as switching to a new technological baseline that may or may not out-perform the old one presents great commercial risk. In this regard it is beneficial to be able to identify early on whether an emerging technology is likely to have scope for development beyond that of the current dominant technology, and commercially, when the tipping point might occur where the new candidate would become the industry ‘mainstream’ technology option.
The growth opportunities presented by Large Technological Systems (LTS), such as the aerospace sector, may also be significant and present another form of risk: Airbus Global Market Forecast identifies the need for almost 20,000 new single aisle aircraft alone by 2030 \cite{RN729}. This is derived from replacement of existing aircraft but also new aircraft to support traffic growth in both established and emerging markets. During the same timeframe, fleet growth is also expected in the twin-aisle and large aircraft markets, bringing the expected value of market growth to be more than $4 trillion \cite{S.A.S2013}.
The size and complexity of this market means that success in securing these new aircraft sales will depend on being able to deliver a product that meets the needs of a wide range of Air Transportation System (ATS) stakeholders, each with different strategic goals. Based on the significant growth trends expected, these forecasts illustrate how there is much to gain in the aviation industry, but equally, much to lose for incumbent firms. As a result, within technology and product development decision-making processes significant emphasis is often now put on risk management practices in order to minimise a firm’s exposure to both technology and market volatility. In the case of technological substitutions it may not be immediately apparent what the benefits and risks of this new technology could be without having a comparable example that already exists to gauge it against. Typically risk is quantified as the financial investment, time, and effort required for the development of the new technology that is lost if the project is not successful, whilst the benefits will most likely be defined via existing 'legacy' performance metrics (at least initially). In aviation, weight and fuel burn have conventionally been the principal design drivers for technology performance assessment, but in recent years alternative metrics that reflect a direct impact on product and service attractiveness to customers (such as the aircraft’s impact on airport capacity constraints) have started to become increasingly important. Whereas the metrics used to characterise the benefits of an emerging technology may themselves evolve over time, the performance expectations that drive eventual adoption of new technologies are often rooted in known scientific principles. Consequently, studies of historical adoption patterns of emerging technologies driven by scientific, rather than commercial, expectations may provide a better insight into how the performance of future innovations will evolve over time in these cases, along with observed market responses. Disruptive innovations also present a greater magnitude of benefit and/or risks in comparison to incremental development strategies, arising from the gamble of being the ‘first-mover’ in a new technological field. On the one hand, the sunk cost of prototyping, development, and industrialisation learning curves associated with pursuing a disruptive technology will no doubt be steep, but on the other hand the rewards for being the first to capture the market can equally be of great significance. Conversely, adopting a low cost, low technical risk, incremental development strategy can actually turn out in some cases to be higher risk than adopting a new technological paradigm if the incremental technologies developed are easily replicated, or even leap-frogged, by competitors. In the worst case, companies may commit extensive resources to technologies and strategies that may be obsolete by the time they come to market. This emphasises a common challenge faced by many technologies, innovations, and business models when first introduced into commercial markets, in terms of the assessment of the projected viability of the product or service being offered in uncertain future conditions. To this extent forecasting techniques are often used to project market outcomes, assisting organisations in determining strategies by providing a guide to future opportunities, risks, challenges, and areas of uncertainty. Forecasts are used in many different aspects of life in order to provide some guidance on the implications of potential changes: from predictions of changing weather patterns, to projections of a nation’s financial outlook, or to provide up-to-date warnings of traffic congestion to in-car satellite navigation systems as holiday-makers converge on popular destinations. As computational power has grown dramatically since early numerical methods, these forecasts are increasingly being based on computer-generated simulations of the world (alongside other non-computational approaches such as those based on industry surveys or expert-elicitation). Equally, computer-generated forecasts are increasingly used to represent the possible outcomes of disruptive changes and events that cannot be easily or safely reproduced through conventional experimentation procedures (such as simulating responses to natural disasters, and large-scale social disruptions). Ultimately, the increasingly complex nature of the Air Transportation System, and other Large Technological Systems (which include technical, economic, cultural, and organisational technology adoption influences), means data-driven models may be helpful in providing further understanding of the impact technologies will have on future market evolution. More specifically, the ability to recognise the mode of adoption early on in the development of a new technology based on historical patterns would provide a clearer view of the long-term commercial potential of that technology. In this regard, advances in pattern recognition and behavioural modelling techniques over the past couple of decades may provide a means to make sense of some of the complex sociotechnical influences behind substitutions in Large Technological Systems. As such, this study attempts to bridge the gap between technology performance expectations, technology development patterns, and market uptake in order to improve the robustness, and consequently likelihood of success, of technology development decisions whilst still in the conceptual stages of design.
The perils of a rapidly changing technology roadmap can be illustrated by considering analysis conducted internally within Airbus examining the impact of Undesirable Effects in Design to Manufacture (a survey across the UK business). This study revealed that consistently more time was wasted in re-planning of work than actually completing the necessary tasks using Airbus processes \cite{RN486}, and that much of this re-planning stemmed from the oscillation between conflicting business directions. This survey identified over 300 different Undesirable Effects (UDEs) in Design to Manufacture from 691 reported observations, and through a cause and effect analysis, identified 22 primary UDEs responsible for this situation (crucially these results and proposed solutions were ratified by members from all parts of the UK business). For wing development programs the most significant standalone undesirable event in design is perceived to cost upwards of £3 million to rectify. In addition the Airbus UDE survey identified that delays associated with immature tools and processes can account for over a year’s worth of additional effort, whilst time wasted gathering and reformatting data and reworking of components can cause delays of more than 5 and 6 months respectively. Instances of component re-work in particular are perceived to account for over £2 million of additional costs during this 6 month period. Finally, the survey also identified that the impact of not working to originally agreed plans can account for an additional 50% of required effort in compensation.
Perception of cost is as important as actual cost for decision-making purposes, as in most cases it is the perception of forecasted costs that is used as the basis of decision-making, rather than already known costs. The results of this analysis suggested that 5% of the UDEs observed cost more than £1,000,000, 6% cost more than £100,000, 12% cost more than £10,000, 16% cost more than £1,000, and 6% cost more than £100. Taking these minimum category values and expected frequency of UDE occurrences, this would consequently equate to over £250 million per year perceived as being spent on UDEs in UK wing design.
Previous historic studies have estimated that 65% of aircraft lifecycle costs (LCC) are effectively ’locked-in’ during the conceptual design stages, with 85% of LCC being ‘locked-in’ by the end of the preliminary design stage \cite{RN487}. This may suggest that 65% percent of these annual UDE costs (i.e. over £160 million) are in some way connected to decisions taken during the conceptual design phase. This is particularly significant when you consider that many aircraft have lifespans of 30 to 40 years (or more) and will undergo numerous modifications during their operational life \cite{RN937}. Consequently, this highlights the need to develop stable views on technology development trajectories at the earliest stage of design to minimise, to as large an extent as possible, these recurring costs arising later on from re-configuration of products, designs, tools, and processes.
Beyond this, in terms of the direct return on investment (ROI) of this study, the case for modelling and simulation is often hard to quantify (as the alternative cases of not undertaking modelling and simulation are not usually measured), however literature exists that goes some way to providing an insight. Pharmaceutical and material science firms, which face similar development programme overheads and timescales to aviation due to the equivalent level of certification and regulation in place, are quoted as receiving between $3 and $9 ROI for every $1 invested in modelling and simulation \cite{RN937,swenson2004modeling}. It is therefore believed that the models developed in this study, if not the end product in themselves, will still provide a benefit to decision-making that will enable similar improved efficiencies.
Research purpose
With this in mind, the overall purpose of this study can be stated as the development of a modelling framework to identify and test the sensitivity of the mode of substitution for emerging technologies in Large Technological Systems, such as the ATS. The models developed in this study are intended to form part of a product and technology lifecycle management toolset, enabling dissimilar technology options (potentially at different stages of development) to be evaluated versus performance expectations and anticipated market responses. Hypothetical product and/or technology ‘roadmaps’ could then be tested against market responses, in order to inform the systematic targeting of funding to bring technology performance to the required levels at or ahead of the expected time, consequently guiding technology substitution.
Research objectives
In order to meet this purpose, specific research objectives are outlined here that this study aims to fulfil, whilst the study hypotheses, research questions, boundaries, desired outcomes, and research strategy are explored in more detail in chapter 3:
- Identification of technology substitution patterns and characteristics in Large Technological Systems in relation to scientific & technological development efforts and other sociotechnical influences
- Identification of historical technology substitution case studies to determine the impact of different patterns of scientific & technological development efforts on global technology adoption trends
- Development of a technology classification model based on features extracted from historical datasets
- Construction and validation of a dynamic technology adoption modelling framework for use in conceptual level design based on identified substitution patterns and case studies
Consequently, the end focus of this study is on developing a modelling framework that is able use data-driven technological development patterns to reproduce observed substitution behaviours for a range of historical case studies and infer future trends. To provide a clearer overview of the main themes and concepts applied in formulating the framework described in this study, an outline of the chapter structure is provided in Fig. \ref{357546} below: