Check out the episode:
You can find the shownotes through this link.
Are you interested in last-mile urban logistics optimisation?
Our debate today works with the article titled Fulfilment of last-mile urban logistics for sustainable and inclusive smart cities: A case study conducted in Portugal from 2022, by Diogo Correia, Cristiano Vagos, João Lourenço Marques and Leonor Teixeira, published in the International Journal of Logistics Research and Application.
This is a great preparation to our next interview with Laetitia Dablanc in episode 474 talking about how current urban experts lack crucial skills to update urban areas.
Since we are investigating the future of cities, I thought it would be interesting to see the readiness of policymakers to manage last-mile urban logistics for better urban futures. This article highlights that stakeholder collaboration and predictive analytics are essential for transforming urban delivery into a more efficient, community-focused operation.
[intro music]
Welcome to today’s What is The Future For Cities podcast and its Research episode; my name is Fanni, and today we will introduce a research by summarising it. The episode really is just a short summary of the original investigation, and, in case it is interesting enough, I would encourage everyone to check out the whole documentation on the WTF4Cities website where you can find the shownotes and more information on supporting the podcast. This conversation was produced and generated with Gemini LM as two hosts dissecting the whole research.
[music]
Speaker 1: When we picture a city’s circulatory system, we usually think of the major arteries, right? The highways, the train lines, the big shipping ports. It feels designed.
Speaker 2: There is an intentional master plan.
Speaker 1: Exactly. The city engineers know exactly where the heavy traffic is supposed to flow. But then you zoom into the capillaries, the last mile of urban delivery, and suddenly that master plan completely vanishes. We aren’t looking at intentional design anymore.
Speaker 2: No. It is the absolute definition of a logistical free-for-all. You have half a dozen different companies sending half a dozen different vans to the exact same apartment building on the exact same afternoon.
Speaker 1: Yeah. And it’s a growing crisis. E-commerce and our collective expectation for instant deliveries are just flooding city centres with proprietary delivery vehicles. They actually account for up to 21% of CO2 emissions in residential areas now. So the central question we are exploring today is can and should municipal policymakers proactively orchestrate real-time 15-minute last mile logistics through a collaborative algorithmic model?
Speaker 2: And we’re drawing this discussion from a recent Portuguese case study titled Fulfillment of Last Mile Urban Logistics for Sustainable and Inclusive Smart Cities. I contend that while the theoretical algorithm is elegant, placing municipalities at the centre of real-time logistics is fundamentally flawed due to empirical evidence showing a severe lack of municipal data, resources, and coordination capacity.
Speaker 1: And I argue that city-led orchestration of shared predictive logistics networks is an essential technologically feasible step to ensure urban sustainability and fulfill citizens’ real-time needs.
Speaker 2: I agree the current model is inefficient. The vans are driving way too far and carrying way too little.
Speaker 1: Because the traditional supply chain uses this disruption with instant replacement model. You order a specialized cable, and a single van leaves a distant warehouse on the outskirts of the city just to drop a single cardboard box on your porch. It’s totally incompatible with modern demands.
Speaker 2: So what is the alternative?
Speaker 1: It’s what experts call logistics 4.0. It’s a proactive decentralized model. The idea is to use shared mobile stock points, small temporary hubs right in the neighbourhood, and fulfill the final leg with smooth transportation, like cargo bikes. The study provides a viable mathematical blueprint for cities to act as the central nervous system of this shared economy network. By using spatial indexing and clustering, they reduce the vehicles in circulation while proactively meeting demand.
Speaker 2: Okay, but a mathematical blueprint is useless without operational reality. When the researchers actually surveyed Portuguese policymakers, the results were sobering. Over 76% of the municipalities explicitly stated they lack the technological tools for real-time visualization. And when asked if they could mobilize resources for a 15-minute response, only about 2% felt they wouldn’t struggle. Building this municipal framework assumes an agility that frankly bureaucratic public procurement simply does not possess.
Speaker 1: I hear you, but I think you’re underestimating the elegance of the proposed four-step algorithm they tested in Lisbon. It automates the heavy lifting. First, it uses DBSCN clustering to group needs within a 1.5-kilometer radius.
Speaker 2: Identifying the hotspots.
Speaker 1: Exactly. Then it uses Uber’s H3 spatial indexing, basically dropping a massive honeycomb grid over the city map to find the optimal candidate spots for those storage points. And from there, it uses Valhalla routing to draw 15-minute delivery isochrones.
Speaker 2: Isochrones being the actual distance a bike can travel in 15 minutes, accounting for streets and traffic.
Speaker 1: So the city shouldn’t be driving the delivery vans. They should act as the air traffic controller. They provide the spatial indexing and predictive routing that prevents the chaotic overlap of private fleets.
Speaker 2: That is a compelling analogy, but an air traffic controller is completely useless if the radar is broken.
Speaker 1: The radar being the data?
Speaker 2: Yes, I acknowledge the sophistication of H three and Valhalla, but you can’t run a predictive density cluster if you don’t know who needs what. Like I mentioned, over seventy-six percent of cities lack detailed citizen data.
Speaker 1: Sure. But some smaller municipalities, the ones under twenty-five thousand people, claim to have this data.
Speaker 2: Yeah, but in a town that size, data just means the local pharmacist knows Mrs. Smith needs her heart medication. It’s personal relationships, not a digitized, standardized database. And larger cities explicitly lack it due to GDPR restrictions. You can’t legally maintain a real-time database of citizens’ dietary and medical profiles without massive privacy hurdles. Without that real-time data, the predictive algorithm has absolutely nothing to process
Speaker 1: GDPR is a real hurdle, but the data deficit can be bridged through a sharing economy open cross-company network.
Speaker 2: Meaning what exactly?
Speaker 1: Meaning the city doesn’t collect everything directly. Almost 88% of surveyed policymakers felt confident they could meet citizens’ needs if they had dynamic cooperation. By utilizing existing private micro-logistics operators and moving deliveries to local storage points during off-peak hours, the city just regulates the collaboration rather than footing the bill.
Speaker 2: I’m sorry, but that’s an incredibly optimistic win-win assumption.
Speaker 1: Why?
Speaker 2: Because 51% of policymakers admitted their biggest dependency would be on private human resources and 27% on private logistics coordination. And the study highlighted 74 Portuguese e-commerce companies lacking capacity for instant deliveries themselves.
Speaker 1: Exactly, which is why they need a shared system.
Speaker 2: But why would highly competitive private entities willingly adopt open data standards and surrender their proprietary routing advantages to a municipal platform? Routing efficiency is their entire moat. It’s how they win.
Speaker 1: Because the current proprietary reality is bleeding them dry too. But look, even if we assume private entities agree to share resources under a municipal umbrella, we have to look at the proactive 15-minute fulfillment model itself.
Speaker 2: Okay, let’s look at the physical realities of that.
Speaker 1: Because by moving away from instant replacement to pre-positioning goods based on predictive needs, deliveries can be made strictly by non-polluting means, like bikes or scooters. This solves the emissions issue and addresses social concerns like population aging by ensuring essential goods are always within a 15-minute reach.
Speaker 2: But a 15-minute isochrone drawn on a map by Valhalla doesn’t account for actual logistical bottlenecks. The policymakers raised huge practical limitations.
Speaker 1: Like what? The bikes?
Speaker 2: Topography, for one. Lisbon is famous for mountainous hills. A 15-minute bike ride on a flat map is very different from pedalling cargo up a 15-degree incline. There are rural accessibility issues and a severe lack of cold storage options for perishable goods or medicines.
Speaker 1: They can outfit local mobile stock points for that.
Speaker 2: But how? A 15-minute algorithm doesn’t account for the high bureaucratic burden of public procurement needed to operationalize intermediary logistics during real-time events. You can’t just throw up a refrigerated shipping container on a sidewalk without contracts, safety checks, and staffing.
Speaker 1: You wouldn’t be doing it on the fly. The procurement framework would be established in advance with pre-approved vendors. The algorithm would dynamically assign resources within those already signed contracts. Yes, it requires overcoming significant data sharing and bureaucratic hurdles, but the study proves the technological framework for a sustainable shared resource logistics network exists.
Speaker 2: So cities just need to try harder.
Speaker 1: Cities must embrace this algorithmic approach. We have to stop the unsustainable sprawl of private delivery vehicles. It’s the only way forward.
Speaker 2: I hear your point. But while the spatial indexing and isochrone mapping present a fascinating theoretical exercise, the severe lack of municipal tools, the strict realities of GDPR, and the inherent friction of public-private coordination just makes city-led 15-minute logistics an impractical solution for the near future. The radar is just too broken right now.
Speaker 1: I think we can both agree that the current proprietary siloed approach to last mile delivery is environmentally and operationally unsustainable.
Speaker 2: Absolutely.
Speaker 1: And it seems inevitable that modern spatial tools like H3 and OpenStreetMaps will play a major role in whatever solution emerges.
Speaker 2: Definitely. The math works even if the bureaucracy hasn’t caught up yet.
Speaker 1: Exactly. So we’ll leave it to the listener to weigh the potential of municipal algorithmic orchestration against the realities of bureaucratic and data limitations. The evolving intersection of smart cities and logistics 4.0 provides really rich material for continued exploration.
[music]
What is the future for cities podcast?
Episode and transcript generated with Descript assistance (affiliate link).
Leave a Reply