Applying the metrics of economic geographer Prof Bent Flyvbjerg, Ireland is far from being a standout failure on megaprojects. <\/p>\n
This may comfort those in the Government<\/a> and State agencies seeking to scale up critical infrastructure as well as private-sector developers and investors hoping to deliver as planned within reasonable time frames. <\/p>\n We live in the age of the megaproject and it is probably the most important and demanding instance where the public and private sectors must collaborate \u2013 and know profoundly their respective roles. But Flyvbjerg\u2019s research shows the combination rarely delivers as required. <\/p>\n A megaproject is a large-scale, complex and often transformative venture, typically costing $1 billion or more, requiring many years to develop and build, involving numerous public and private stakeholders. <\/p>\n Flyvbjerg realised trillions of dollars were being spent on megaprojects, yet \u201cnobody had reliable data on how [they] were performing\u201d, he told an online gathering hosted by the clean energy think tank Trifecta Ireland recently \u2013 it was staged to discuss how Ireland might ramp up delivery of renewable energy projects (some of which are megaprojects) in line with ambitious Government climate<\/a> targets. <\/p>\n Addressing the immense information gap led to the 2023 book How Big Things Get Done (written with Dan Gardner). It has yielded a vast amount of data arising from 16,000 projects. For most stakeholders, the associated analysis is chastening. <\/p>\n There are \u201cthe famous 0.5 per cent\u201d; the projects that come in on budget, on time and deliver the promised benefits or better. A mere 8.5 per cent in their vast database met cost and schedule targets. <\/p>\n Through work conducted by his team, Oxford Global Projects at Oxford University<\/a>, <\/b>decades of research and experience in dealing with megaprojects \u2013 what works; what doesn\u2019t \u2013 is now available. <\/p>\n It might seem many learnings are counterintuitive, or even simplistic, given the complexities of megaprojects but the same core messages keep surfacing. The book includes 11 heuristics for better project management; \u201cfast and frugal rules of thumb used to simplify complex decisions\u201d. <\/p>\n Some range across \u201cthink slow, act fast\u201d; know the biggest risk is you (your behavioural biases) and drive out complexity while scaling up modularisation. It includes early experimentation and, appropriately for a Dane, building with Lego<\/a> (metaphorically), the basic building block. <\/p>\n \u201cThat\u2019s how a single solar cell becomes a solar panel, which becomes a solar array, which becomes a massive megawatt-churning solar farm.\u201d So \u201cwhat\u2019s your Lego?\u201d is asked. <\/p>\n These principles should never be used like \u201cthoughtless paint-by-number rules\u201d. They must be adapted to requirements of particular megaprojects but come with the backing of what are now indications from 20,000 projects. Trifecta Ireland, a not-for-profit, is planning to shape an energy master plan for Ireland incorporating Flyvbjerg thinking. <\/p>\n Above all, he says, avoid \u201cthe fat tail\u201d \u2013 the deadly mix of large budget overruns and time delays. If locked into a pattern of \u201cregression to the tail\u201d, it is only a matter of time until a new extreme event occurs, \u201cwith an overrun larger than the largest so far\u201d, and thus more disruptive and less plannable. <\/p>\n His team analysed 25 different project types. The fat-tailers are IT projects; nuclear storage, defence, Olympic Games, nuclear power and dams, where overruns of up to 600 per cent are not unusual. <\/p>\n \u201cYou need to know whether your project type is fat tailed or not, and, if yes, how much and how to mitigate the risk. Then you ask \u2018Can we afford the risk?\u2019 and if not, \u2018Should we walk away or can we reduce the risk?\u2019\u201d Look no further than how the UK\u2019s Sizewell C nuclear plant has ballooned in cost to \u00a338 billion from a \u00a320 billion figure in 2020. <\/p>\n \u201cBlack-swan management\u201d to mitigate high-impact events may be needed to cut the tail. A simple example is Fukushima, he says. <\/p>\n \u201cYou need to protect the nuclear power plant against a tsunami wave. Simple. You build a wall that\u2019s tall enough.\u201d <\/p>\n Fortunately, Flyvbjerg notes renewables<\/a> have the thinnest tails, headed by solar, energy transmission (grids) and wind \u2013 and probably long-duration battery storage. And hey \u201cthat\u2019s exactly what we need in order to solve the climate crisis\u201d. <\/p>\n Large systems work best when they grow from smaller, proven parts. Modularity reduces complexity and risk \u2013 a vital approach for scaling up solar, wind farms and grid upgrades, he adds. <\/p>\n Delivering at the pace climate goals demand means designing projects and programmes that can scale easily, knowing necessary \u201csuper scaling only happens with exponential growth\u201d. If something can\u2019t scale, we need to rethink it. <\/p>\n Engineers \u2013 and others \u2013 are prone to \u201clock in\u201d. They focus on particular solutions without considering alternatives. \u201cWe like to focus on one thing quickly. It simplifies our lives but it creates a lot of problems later, if you anchor on the wrong thing … We tend to think linearly, our brains are good at that and extremely poor at thinking non-linearly and probabilistically.\u201d <\/p>\n With relentless focus on speed of delivery, where modules are built in factories and sites are for assembly, it becomes possible to install a \u20ac1 billion wind farm in less than a year, he suggests. <\/p>\n His most stinging criticism is of the construction sector\u2019s failure to improve productivity. Buildings are invariably bespoke, unlike modularisation of the car, which brought highly efficient and cheap vehicles. <\/p>\n He cites factories in China<\/a> building homes, schools and hotels and then bringing them on site for assembly. \u201cIf I worked in construction industry, I would be scared of what they\u2019re doing in China, because just like with cars, once somebody gets a handle on increasing productivity in a manufactured way in construction, anybody who can\u2019t do that would die, just like anybody who couldn\u2019t do it in the car business has already died.\u201d <\/p>\n Electricity distribution has a similar problem but is solvable, Flyvbjerg believes. Once co-operation goes beyond borders and countries realise who their neighbours are in the broadest sense, transformation including a supergrid is possible, provided there is an international market for electricity. This is key to cheap electricity for Ireland in particular \u2013 and modularisation is doable here, too. <\/p>\n Inevitably, he is asked about the National Children\u2019s Hospital<\/a> marred by delay and budget overruns. Denmark<\/a> is in a similar bind, he says, building seven new hospitals all at once. Both cases are notable for lack of standardisation, failure to learn from other hospitals and \u201cuniqueness bias\u201d. <\/p>\n \u201cI bet you that in Ireland you tend to see this hospital as unique. We haven\u2019t built a hospital like this before, and it\u2019s Irish, so there\u2019s not much we can learn from other hospitals, and certainly not other hospitals in other countries … So you\u2019re reinventing the wheel over and over and over again in project management and in construction, because of this unique bias where you think that, \u2018Yeah, this project is different\u2019.\u201d <\/p>\n He is a big advocate of \u201creference-class forecasting\u201d in both the public and private sectors; a method used to make more accurate predictions by drawing on data from comparable past projects. It aims to mitigate effects of optimism bias and other forecasting errors by using an \u201coutside view\u201d based on historical data, rather than relying solely on specific details of the project at hand. <\/p>\n Ireland is not behind the door in applying it, he adds. Transport Infrastructure Ireland<\/a> was an early adopter, as are the National Transport Authority<\/a>, Gas Networks Ireland<\/a>, Uisce \u00c9ireann<\/a> and some hospitals. It has also been applied to MetroLink<\/a>, Dart+<\/a> and BusConnects<\/a>. <\/p>\n He would not favour short-circuiting planning and permitting: \u201cI\u2019m not sure that\u2019s desirable. But there are examples of governments taking a different tack on this and saying, \u2018This has to happen, we have to do this\u2019 … so get the permit process geared to this and make sure that it\u2019s done properly but fast; not the usual bureaucracy.\u201d <\/p>\n