[MUSIC] What if we could predict the future, wouldn't that be great? But sorry, we can't, but we can make some quite qualified guesses and that's what this module would cover. When you have completed this module, you will be able to use one particular method to build the model of the future, your map of the future landscape. But as we discussed before, we only have data about the past but not about the future. But there are a number of methods to build models of what the future might look like. We use these models to take better decisions and to prepare ourselves for what may happen. Why do we do scenario planning and foresighting? Well, that's due to that the future is not predetermined or predictable. And of course, if it was, I mean, there would be no point of taking action today because it would not have any effect at all on the future. And also the reason is that we do not have all the information about the future, so we need to have some help to make these guesses. These types of method had been used since the 1970s and since the successful development implementation of the company Shell. But the desire to predict the future has been there for thousands of years, as well as the fortune tellers. When we build a model or a theory, we use historical data and with using induction, we produce this model or theory. Once we have the model of theory, we use that to make some prediction of what may happen in the future. And that's then the deduction part of this process. How good is this then? Well, I said foresight is not prediction. It's about getting an idea what plausible futures might look like, but we have a number of issues here. There is a lot we don't know. We have the historical data and we also know that the future is, there are some questions to ask, but the main issue is that the things we do not know that we do not know. So there are a number of questions that we will never ask ourselves because we don't know that they need to be asked. What different futures do we have then? Well, we have possible futures. In the future, kind of anything is possible. Anything could happen, we don't know so much or actually nothing about the future, though there's a number of guesses. Then we have plausible futures that could happen. There is a logical link to that this may happen that we can understand. Then we have probable futures, and that's likely to happen and then maybe the time horizon is not far into the future. And then we have desired futures, futures that we want to happen. In this course, we looked at how technology made the impact of technology and there we are somewhere in between, I would say, probable and plausible of future. In this graph, we see how the uncertainty is increasing fast when we try to predict what may happen in the future, and also the predictabilities then going down as a consequence. When we look into, I mean, just the coming few years, we usually call that forecasting. That's using the present trends that we have and it's kind of likely if something is growing 20% this year, it may grow 20% next year. But that, of course, has a limit to how far into the future we could use that. And then we come to the foresight methods if we have a longer time horizon. And if we go very far into the future, it's just a bunch of wild guesses and uncertainty is very, very high. We have a number of different methods to use, but when to use what method? Well, it depends on your time perspective. How far into the future would you like to look? But it's also a question of how you intend to use the model. If it's part of a risk assessment, it may be a good idea to work with a number of plausible solutions. But if you need a joint map to base a common decision on, it may be a better idea to work with just one scenario, and that's what we're going to do also in this course. If we look at the classical scenario methodology is based on two key uncertainties. And depending on how these key uncertainties develop, we get four different plausible futures. This classical scenario planning methodology was first developed by the company Shell in the 1960s. In 1973, the world was shocked by the oil crisis, but Shell wasn't. They had managed to use this methodology to prepare themselves for that an oil crisis may actually happen. So they had already a plan for how to act when that oil crisis would arrive. Back-casting, in the classical back-casting methodology, the starting point is usually a desired future, and the question is how we should act to make sure that the desired future materializes. This methodology was derived around 1990 and it's mainly being used within the sustainability area. In the sequential back-casting that we'll use in this course, the starting point is not a decide future, but what a future would look like if we fully utilize the technology that we have today. Is this not a oversimplification then, just using one scenario, that may go quite wrong? Well, yes, that's true, but at the same time you need to remember that a model is always a simplification. And we need to have a balance between how complex we build the model and what the model should be used for. Working with one scenario makes it very simple, and with simplicity we also get usability. So in that case, we get something that can actually be used by people. In this module, we will focus on introducing the sequential back-casting. And we will run a type of interactive workshop with you, as interactive as it now can be doing a MOOC, but anyhow, it will illustrate what the impact may be on society. And the technology we'll look at, that's self-driving cars, that's the case we will follow when doing this sequential back-costing. So looking forward to doing that with you, and let's see what happens.
