Computing the Impossible #TechVision Look at the world of logistics, for example. At the While this wouldn’t eliminate the need for highest level, the logistics industry can be boiled logistics companies, making a central problem down to one question: How do we move things readily solvable would fundamentally transform through multiple locations most efficiently? what they do and how they do it. Consider the 69% This is also known as the “traveling salesman” improvements that would come with dynamic problem, which is considered computationally optimization of last-mile delivery in any city of global executives report intractable due to the sheer volume of possibility. with any fleet size – not only could heightened quantum computing will have A single trip with just 16 stops has 20 trillion consumer expectations for near-instantaneous a breakthrough or transformational possible routes, and a classical computer would delivery times be met, but traffic congestion and impact on their organizations need to compute every single possibility to environmental impacts could be dramatically in the future. find the most efficient one – an effort that even reduced. And on a much larger scale too, the fastest computers could take decades to worldwide freight transport efficiency would 170 compute. Efficiency is the table stake for see massive improvements and enable the most logistics companies, but solving this core accommodation of sudden changes through efficiency problem is nearly impossible. So adaptive asset reallocation. instead, logistics companies solve subsections of the problem: tendering goods from shippers Every industry has “grand challenges” that directly to carriers, joining consortiums to lower define industry boundaries and inform the costs, and more recently relying on the power of core products, services, and strategies within. cloud and data to improve routing and tracking For instance, pharmaceutical companies are services. They are competing around the currently constrained by the protein folding traveling salesman problem – never solving it. problem, and the financial services industry Routes become better but never optimal. as we know it is predicated on the assumption that predicting the stock market and accurately Like with security encryption, the traveling modeling risk are very hard to do. But what if salesman problem could be solved by a they weren’t? powerful enough quantum computer. Introduction // WebMe // Programmable World // The Unreal // Computing the Impossible 75