Amazon has long threatened traditional logistics companies with its advanced predictive analytics capabilities, sometimes even shipping products before customers place their orders. And now, legacy logistics companies are fighting to catch up. UPS, for example, is in the process of implementing a new predictive analytics tool that analyzes over one billion data points daily to estimate capacity, package volumes and customer demand.
“If we don’t do our job well, then there’s no doubt that big, strong companies like Amazon will look into whether they can do better themselves,”
Soren Skou, Maersk CEO
Even the world’s biggest container shipping line is worried about Amazon: “If we don’t do our job well, then there’s no doubt that big, strong companies like Amazon will look into whether they can do better themselves,” said Maersk CEO Soren Skou in an interview with Bloomberg earlier this year, referencing the e-commerce giant’s plans to begin handling more of its own deliveries.
As a logistics company, you know it’s hard to predict the future. But these events should tell you one thing for certain: If you don’t start using a predictive logistics solution today, you won’t be around for tomorrow. Such predictive analytics technologies have no shortage of use cases: they can be used for network management, risk management, demand and capacity planning, predictive maintenance, route optimization, and more.
So, while there’s no longer any question about the importance of implementing a predictive analytics solution, what should you do to prepare?
Digitize processes and change mentalities
For starters, the transportation and logistics industry as a whole needs to step up its game. According to Gartner’s 2017 CSCO survey, over three-quarters of chief supply chain officers admit that their digital transformation projects are still not aligned. And it’s really no surprise; the lack of the penetration of the latest technology in the industry leads to using such tools as Excel and even paper-based records – neither of which are particularly useful when it comes to performing advanced analytics or forecasting.
As such, the first step you should take before investing in a predictive analytics solution is to put a formal plan in place for the digitization of your company. In some cases, this could mean hiring or appointing a Chief Digital Officer to come on board to guide the digital transformation of your business and build an information-driven supply chain. This person could also be responsible for laying out a plan to implement technologies like sensors, GPS and other IoT devices across all of your company’s facilities and equipment, which would then generate a huge amount of raw data for analysis and help guide business decisions.
As the logistics sector is highly competitive and operates on extremely thin profit margins, recording this kind of data that gives a birds-eye view over the entire supply chain – much in the same way as Amazon does – is the only way to stay relevant and competitive.
However, your employees must also be on board with this new technology and have the skills to use it. Offering them the opportunity to attend industry conferences to become familiar with new technologies, or training sessions to learn how to use them is, therefore, another critical element to consider.
Clean data for effective analysis
Apart from recording data in the first place, another major consideration is how to prepare that data for analysis. In other words, just because you have data doesn’t mean you can suddenly make valuable business decisions. In fact, a Deloitte study in 2017 found that data quality was the main barrier to the effective application of digital technology in logistics organizations for almost half of chief procurement officers surveyed. Lack of data integration was the second biggest challenge.
Again, these findings are no big surprise. Logistics companies often record data in different systems, and different units of measurement, which makes effective data analysis almost impossible. With this in mind, it’s important to try to standardize processes for recording data in your organization. For example, will you record everything by volume, size or weight? (Hint: Individual dimensions, that is, height, width and length, are the best for data analysis, but it’s one of the hardest measurements to capture. Volume is the second best, followed by weight.)
Additionally, in many cases, recorded data is simply inaccurate. Maersk, for example, reported in 2013 that nearly 12 percent of container industry invoices are inaccurate, with other industry players putting the error rate as high as 30 percent. Needless to say, these data errors can cost logistics companies millions of dollars per year.
Fortunately, AI can help clean up data before it is used for predictive logistics – both in the sense of standardizing it, and identifying and filtering out errors that otherwise skew the results. Such data cleansing is arguably the most important step to prepare data for predictive analytics. And AI can help not just with cleaning up the data, but also with enriching it by predicting missing values. Interestingly enough, we’ve found that it only requires five to 10 percent of accurate, complete historical data about a company’s shipments or inventory for AI to build a full data set that is ready to be used for predictive purposes.
Invest in the right people and technologies
Once you have systems in place to record and clean data, it’s time to consider what investments you can make to best take advantage of it. Currently, most logistics companies rely heavily on people’s years of experience to make predictions about demand and capacity. And while such expertise is undoubtedly valuable, you need to have the right technology to complement this expertise. In fact, the World Economic Forum estimates that technologies such as data-driven information services could generate an $810 billion upside potential for the industry.
But, building such tools from scratch is far from the core focus of most companies, and only within the budgets of the giants like FedEx or UPS. This is why a number of innovative startups are now focusing on using artificial intelligence and machine learning to help companies forecast demand, predictively manage assets, optimize routes, reduce maintenance costs and improve utilization through data analytics. For companies without the resources to develop such technologies internally, exploring the tools available on the market is the most viable option.
Obviously, such technologies can bring value to your company, but the most positive outcome can only be achieved when your people and your technologies are synergistic. If it’s in the budget, this means hiring data experts or upskilling existing employees to perform some of the manual data cleansing tasks and to analyze data for important trends or errors that might have gone unnoticed by the technologies. For most companies, though, this means looking at a reliable provider such as Transmetrics that can do the same at a relatively lower cost. In doing so, your organization will be well-prepared to reap the benefits and insights of the data at your disposal.
At the end of the day, investing in a predictive analytics solution is no longer an option – it has become a necessity to maintain competitiveness with Amazon and the other logistics giants that are pushing ahead in adopting emerging technologies. With that in mind, the time is now to invest in your people and your technologies to bring your business into – and help it thrive in – the 21st century.