The data behind teamwork
It may be hard to remember now, but there was once a time when online order and delivery wasn’t the default option for most of our purchases.
That was before lockdowns accelerated the transition away from in-person shopping, though, and like so many societal shifts of the past two years, it seems likely to stick.
The downsides of this shift are well-documented: struggling small businesses, surging delivery traffic, and above all, the human cost borne by legions of delivery workers. Stories of underpaid, exploited gig workers hustling to make their daily quota while racking up parking tickets and injuries are commonplace. The old guard of home delivery—UPS, DHL, FedEx, national postal systems—seems positively worker-friendly compared to the plight faced by contractors working for high-volume e-commerce retailers. But customers’ expectation for fast, cheap delivery is essentially fixed at this point, and so is the gig economy that fulfills it.
Every now and then a tech fix hints that it might lead us out of this mess. But so far it seems unlikely that any combination of drones, automated lockers, and self-driving vehicles is going to make a serious dent in the problem for several years. In the meantime, some forward-thinking designers and a number of established transportation companies have been looking for solutions from another angle: the data angle.
We tend to think of delivery as a physical problem—putting boxes into vans and moving vans down the road—but any courier, dispatcher or station manager can tell you that it’s just as much an information problem. Where each package needs to go, who’s going to take it, in what order and by what mode is a fantastically complex equation, with huge costs (which often trickle down to the couriers) if you get it wrong.
Why, for instance, are so many urban deliveries made by van, even where it’s difficult, time-consuming, and often illegal to park them? Because some deliveries need a motor vehicle, and it’s often easier to put all the parcels into one van and send it into the streets than to figure out which ones are better delivered by bike or on foot.
But what if it wasn’t? What if there was a way to plot out the destination and size of every parcel to be delivered in an urban area and quickly create an optimal plan for getting them all to their destination in the least costly, least stressful way possible? For this scenario to work, vans would still move large volumes of parcels partway along their journey, then split them out to couriers on foot, bike or other small vehicle, who can skip the traffic jams and parking challenges that make so much misery for the van driver.
If you’ve followed advances in machine learning over the past few years, you might already see where this is going. Figuring out where and when parcel transfers happen, who carries what, and what path they take next is exactly the kind of big, multi-variable problem at which modern algorithms excel. It’s perhaps no surprise, then, that several big companies are already using this approach to tackle the “last mile” problem in-home delivery.
One of the most promising is Ford Last Mile Delivery, an initiative that’s been trialed successfully in both London and Manchester over the past year. Ford’s Mobility Innovation studio teamed up with Smart Design to develop the first prototype of a software platform called Mode:Link, which sets urban meetup points where delivery vans transfer their parcels to four-person teams of cycle couriers and pedestrian porters. Algorithms dynamically update the meeting points throughout the day, based on parcel destinations, real-time courier locations, and live traffic patterns, then communicate this to the entire team through smartphone and web apps. And since then, the platform has developed further to include parcel sorting and enhanced routing capabilities that support these kinds of multi-modal deliveries.
In trials, Ford’s software has shown the potential to dramatically increase daily deliveries per van by shifting much of the load to other modes, while reducing parking and traffic headaches. This is an obvious big win for delivery and logistics companies, since it improves the efficiency of their delivery process while reducing vehicle fuel and maintenance costs. It also reduces emissions and gets motor vehicles off the road, especially in high-density urban areas—by some estimates, delivery vans currently account for over 20% of urban vehicle trips in London, yet traffic crawls at an average speed of 7mph.
Most of all, though, logistic solutions like Ford’s software are a breakthrough for the couriers on the front lines of home delivery. In conversation with delivery workers for the Last Mile project, Smart researchers found that their biggest sources of stress weren’t apartment stairs or heavy boxes, but the costs and difficulties associated with driving and parking. A multi-modal solution removes those issues completely for four-fifths of the delivery team and reduces them considerably for the other fifth who do the driving.
Ford Last Mile Delivery isn’t the only foray into smarter multi-modal delivery systems, of course. Big logistics companies have been tinkering with ways to streamline the last mile for several years now. FedEx, for example, currently has several initiatives in place to improve efficiency through better tracking, reduced hand-scanning, better analysis and—yes—autonomous delivery. DHL is using a combination of electric vans, boats, and bicycles to deliver parcels in the most congested parts of central Amsterdam. And numerous urban centers throughout Europe have started banning conventional delivery vans, prompting private companies to develop multi-modal solutions out of necessity. This transition is helped along considerably by the safer, more efficient bicycle infrastructure that many of these cities have spent years developing.
That the most promising fixes to the delivery problem come from smarter data crunching rather than physical gadgets shouldn’t be that surprising. This is the era of big data, after all, and that means more than just e-commerce companies analyzing your browsing habits. It also means knowing where things are in the physical world, and what’s going on around them. The current global logistics crisis—from backed-up ports and stretched supply chains to expensive lumber and empty shelves—now looks like it’s going to be solved through AI-powered optimization rather than more ships or truck drivers.
The world we grew up in was rife with inefficiencies and clunky workarounds, but now we’ve got the ability to close those gaps. And while “more efficient” might conjure up images of dehumanized assembly lines and rigid corporate hierarchy, it can also mean vans that aren’t stuck in traffic, and parcel couriers who know exactly where to go next in order to finish their route on time. That kind of efficiency isn’t just a benefit in economic or resource terms. It’s also, in some important ways, how we adapt to our modern, hyperlinked world without destroying the people who make it possible.
About Jasper Dekker
Jasper is an associate design director who strives to make technology meaningful, useable and delightful for all people. Next to his UX/UI skills, he brings expertise in embedded UI and design systems and has worked across automotive and mobility, consumer tech, media, healthcare and gaming. His notable clients include HP, Ford, Amplifon, Google, Gatorade, Merck, Samsung and his work has been awarded by IDSA and Fast Company. Outside Smart, Jasper is an advocate for urban cycling and dabbles in creating hifi gear.
Smart Cities World
How design can pave the road towards a new personal mobility
Smart Cities World
Three ways autonomous vehicles will impact families
Virtual Smart Salon
The road to well-being: Designing inclusive and thoughtful mobilit...
Last mile delivery
Do bike lanes actually make cyclists safer?
Using data science to analyze accidents in 3 cities yielded some surprising results.