Our supply chains are the lifeblood of global business. You probably don’t need to be reminded of the stakes involved when companies engage in the movement of products throughout the world ” especially if you deal in time-sensitive merchandise like foods, beverages or health care products.
Big data, data analytics and machine learning, are bringing enormous changes to the transportation, shipping and supply chain industries, all of them for the better. Here’s a look at what’s changing.
1. Demand Forecasting and Facility Planning
The days when companies didn’t have the tools to make informed expansion decisions are far behind us. Company business decisions have to harmoniously balance a vast amount of information and a huge number of variables. That’s where artificial intelligence enters the mix.
Companies hoping to broaden their reach by establishing new routes or breaking ground on distribution centers increasingly call upon their databases of structured and unstructured information. This includes past transaction records, customer data such as location, preferences, purchase history and anticipated future purchases, and the flow of products across shipping routes, as well as vendors and other world partners.
This web of information is increasingly a vital part of choosing business priorities and planning for the future. Knowing where your products are most popular, and at which times of year demand is likely to rise, can all go a long way toward singling out a new city or region for expansion.
Plus, gathering data from aftermarket parts manufacturers or repair shops can shed light on how customers interact with their products and might help you uncover opportunities to expand your service offerings.
2. Connected Infrastructure for More Proactive Maintenance and Operational Awareness
It’s difficult to understate the importance of operational awareness. For warehousing, shipping and transportation companies, success rests in part on having a robust infrastructure and the means to keep track of all of it at once.
The Industrial Internet of Things (IIoT) is the web of connectivity that unifies assets like a facility’s HVAC system, robotic sorting mechanisms, electric forklifts, delivery trucks and much more. Each of these pieces of equipment serves a vital function, which means downtime must be minimized practically at any cost.
Accumulating real-time data about the performance of electronic and mechanical assets means logistics companies can deploy maintenance personnel in advance of a critical failure. In turn, this means business doesn’t have to halt when one of your trucks is due for a tire rotation, or a sensor placed inside a conveyor belt has flagged a suspicious vibration.
If each of these assets is like an appendage for your business, then unifying them with the IIoT is like installing a central nervous system. This is something that 49 percent of companies with fragmented supply chains struggle with.
3. Analytics for Intelligent and Efficient Freight Stowage
Most of us can clearly picture the densely packed inner recesses of a warehouse or the many-stories-high stacks of shipping containers atop seagoing freight vessels.
This shouldn’t come as a surprise, but getting all those containers, case packs, cartons and pallets stowed in the right place, in the right order and under the right conditions is a colossal undertaking. Thankfully, big data, analytics and a touch of machine learning make it far easier to keep things straight.
Warehouse management systems today perform a larger portion of the planning and freight decision-making than ever, thanks to intelligent automation. Robotic sorting in warehouses reduces human error, for example.
Instead of playing it by ear when incoming freight arrives, these semiautomated warehouse management systems and sorting machines call upon all the available data about incoming SKUs. They flag appropriate stowing locations in the facility, according to weight, size, perishability and more.
Of course, stowing incoming freight is just one piece of the puzzle.
4. Machine Learning for Next-Level Order Picking
The same thing can happen in reverse: Order picking in warehouses increasingly leans on automated prompts from management systems for directions to appropriate pick locations, help in maintaining FIFO (first in, first out) and other tasks that ensure employees are working efficiently. It ensures that any specific requirements for product handling are accounted for.
Natural language processing is one branch of machine learning that’s making order picking more accurate. Instead of reading picking directions off a scanner or screen, warehouse employees can give and receive voice-based feedback. They’re directed to the correct location and then can read off a portion of the product number.
The warehouse management system then checks this number against the product it expects to be in the bin, thereby cutting down on incorrect picks and shipments. Called pick-to-voice, a system like this elegantly balances machine intuition with human effort.
5. Automated Inspection and Defect Detection
IBM and other companies are working at an impressive clip to bring greater machine intelligence into processes that previously required human intuition and judgment. One of the most important, from a transportation and logistics standpoint, is the process of carrying out regular inspections and kicking out defective products before they move on to their final destination.
We’ve already seen some compelling demos of AI-driven equipment that’s capable of finding and flagging subtle defects in multiple parts or products.
Today, the threat of faulty, counterfeit or compromised parts is a bigger worry than ever, due to a global economy and electronics and machine components changing hands more regularly than ever. The benefits of bringing AI into this critical task isn’t simply a matter of efficiency ” easier inspection and defect detection through AI stands to greatly improve our collective safety.
6. Artificial Intelligence for a Slimmer Workforce
A lot of the technologies we’re describing here were designed to remove the burden of simple, repetitive tasks from human workers’ shoulders. One of the most open-ended questions in the fields of transportation and logistics right now is just how successful those efforts will be. In other words, how many jobs will AI cost us in these human-centric disciplines?
Based on studies conducted by McKinsey, as much as 49 percent of paid human activity could be ripe for automation through AI. What about logistics, specifically? A perhaps more realistic estimate says that robotics and automation will displace less than 10 percent of human jobs. The nature of fleet management, logistics and material handling processes points to why.
Although these industries will certainly shed some lower-skilled jobs, it will create many other medium-skill positions ” such as monitoring fleets of autonomous vehicles or performing maintenance on robots and cobots.
If big data and automation are bringing great change to your line of work, you’ll have to navigate talent shortages in the computer sciences, like your rivals and downstream partners are. Expertise in data systems is a requirement that’s here to stay in these industries.
An appealing side-effect of replacing a measure of human effort and intuition with the machine variety is that workers can be expected to become more productive and effective in tasks that require empathy, personal contact and other higher order skills.
