In theoretical computer science, the traveling salesman problem asks the following question: "Given a list of cities and the distances between each pair of cities, what is the shortest possible route that visits each city and returns to the origin city?" Anyone who has worked in transport logistics or transportation management knows that in most cases there is no easy answer to this question, and that finding the optimal route between different cities or even different stops along the same tour can be a serious logistical challenge—one that requires planners to manage customer delivery windows, anticipate traffic patterns, and optimize time and distance.
Let’s say you’re moving into a new home. After a long day of loading boxes and furniture into your car and driving them to your new place of residence you’ve finally transported everything into your new house and you’re almost ready to start unpacking and get the place settled. Before you get down to the work of opening up all of your packed boxes, you realize that you haven’t eaten all day, and should probably whip something up before you do any more manual labor. It should be pretty easy to find your kitchen supplies and snack foods, because you made a list of what was in each box before you moved them. One problem: you don’t know which box the list is in.
These days, when most people think of automation, one of their first thoughts is of self-driving cars. What many people don’t realize, as they picture themselves magically napping away their morning commutes, is that when it comes to autonomous vehicles there are actually six levels of autonomy. At level zero, you have a standard automobile, which requires the driver to make every decision and maneuver. At level five, the car itself makes and carries out all of the decisions without any human intervention. In between, we find cars that can maintain speed and avoid other cars on the highway, cars that can change lanes and make turns unassisted, and cars that can perform automated interventions in crisis situations like potential spin-outs.
In the past five to 10 years, real-time information has become a key value-added proposition for bolstering efficiency and decreasing waste in modern, digital supply chains. Businesses have used it to power more agile, responsive processes within their own value streams, creating environments that are primed for improved data-quality and easier analytics integration. The question remains, however, is this technology being utilized to its maximum effect, or are there still use-cases for real-time information that most businesses are failing to fully leverage? The answer is resoundingly the latter, as evidenced by these four surprising uses for real-time supply chain data.
New technologies—from steam power to modern computers—have been the driving forces in supply chain management since the beginnings of industrialized society. Today, supply chain technology is changing at an exponential rate, providing supply chain planners with possibilities that would have seemed like science fiction even a few decades ago. If you follow the news and trends in SCM, you’ve no doubt noticed that machine learning (ML) is often touted as the next major innovation in this long line of technological evolutions—but what, exactly, is it, and how can supply chain managers put it to use?
Let’s say you and a coworker are attempting to find areas of waste in your supply chain. You have a large conference table on which you’ve laid a file that contains all of the transport plans utilized by the company for the past few years. When your coworker hypothesizes that a different grouping of goods would improve fuel efficiency, you need new documents with additional information, meaning that you have to leave the conference room and descend to the basement level where the files are kept. By the time you’ve returned, a new idea has occurred to your coworker, and you have to make a new trip to wherever your files are stored in order to retrieve the necessary information. The result of all this walking to and from the files? Some good cardio, but no plan to speak of.
Each year, topics like big data, advanced analytics, machine learning, and artificial intelligence dominate conversations about supply chain technology, becoming the focus of increasing amounts of speculation as the real power of these technologies becomes clearer and more apparent. As a result, it can be difficult to parse the jargon from the meaningful discourse and the pragmatic analyses from the wishful thinking when deciding how to set expectations for the evolving global supply stream. In the spirit of helping to put all of the hype into context and give supply chain managers the tools they need to look towards an ever-uncertain future, here are a few predictions for the future of advanced analytics:
Imagine you’re a trader on the floor of the New York Stock Exchange. Every morning, you check the prices of the stocks that you’re interested in, and you act on those numbers, not checking them again until the end of the day. Your competition, on the other hand, is using real-time information to inform their trading decisions. Which technique seems more likely to yield a profitable trading strategy? Your knee-jerk reaction is probably that you’re going to lose money virtually every day, because your competition has a more accurate picture of the real financial landscape while you’re using information that’s obsolete virtually as soon as you set foot on the trading floor.
Imagine a world in which trillions of individual pieces of information are gathered each day to create complex predictions about future supply chain disruptions and events. Extremely granular data on trade markets turns information about the movement of goods and services throughout the globe into cognitive systems with the power to illuminate new possibilities and intelligently predict changes in demand before they occur. While this may sound like science fiction, it’s increasingly becoming a reality as supply chains become more and more integrated with machine learning, artificial intelligence, and big data analytics. By 2020, IDC predicts that 50% of supply chains will utilize advanced analytics and artificial intelligence, and the effects on the global supply chain are sure to be widespread.
Let’s say you’re driving down a winding country road to some remote destination. At first, navigating is easy, but as the sun goes down and your headlights come on it becomes more and more difficult to make sure that you’re driving safely down the correct course. Eventually, night has fallen completely and your headlights provide the only illumination—you have to slow your driving speed, so that if an animal or other unexpected nocturnal being wanders into the path of your headlights you’ll have enough time to stop the car. If there’s stormy weather, the visibility becomes even more limited, and the possibility of an unexpected snafu increases.