Area: Smart Cities · TRL: 3 of 9 (proof of concept on paper) · Value-engineering target: USD 120 (estimated cost USD 116) · Difficulty: 2 of 5
A curbside loading bay occupancy sensor for streets that are not plowed: it shows delivery drivers which bays are free and gives cities data on curb use.
Exploded render · Detail render · Interactive 3D model · Concept blueprint (PDF) · General arrangement (PDF) · Sizing calculations · Prototype build plan · Design decisions · Review note
Knowing bay availability reduces circling and double parking. LoadZone puts a low, battery-powered puck in the middle of each vehicle slot of a loading bay. A magnetometer senses the steel body of a parked vehicle, and a LoRaWAN radio from the lab's FieldNode core reports "free" or "occupied" in about 18 s (39 s at worst). A server turns these reports into live bay state for drivers, and occupancy and dwell times for the city; a solar e-paper sign on an existing pole is an option.
It is open and garage-buildable because curb data should belong to the street, not to a vendor. A magnetometer cannot see faces or number plates, so privacy holds even if the firmware is changed, and anyone can inspect how a "free" or "occupied" call is made. The puck is a cast dome over potted off-the-shelf parts, bonded to the road like a raised pavement marker, and the data can be published in the open Curb Data Specification format so cities can compare it with carrier data.
Loading bays are full, and often not with deliveries. In a study of downtown Seattle, passenger vehicles were 52 % of all vehicles parked in commercial load zones, and 41 % of commercial vehicles parked in unauthorized locations, rising to 55 to 65 % near retail centers (Urban Freight Lab, 2019). The same group reports that commercial load zones often run at 90 % occupancy or more during business hours (Urban Freight Lab). A driver who finds the bay taken stops in the traffic lane or the bike lane.
Pressure on the curb keeps rising. Online sales rose from 16 % to 19 % of all retail sales in 2020 (UNCTAD, 2021), and in Great Britain van traffic grew 2.3 % to 59.0 billion vehicle miles in the year to September 2024 (Department for Transport). Cities cannot manage bays they cannot measure, and most loading bays today are measured only by occasional manual surveys.
| Industry | Use |
|---|---|
| Municipal transport and curb management | Occupancy, turnover and dwell time by bay and hour, to size bays and set hours and pricing |
| Parcel, food and freight carriers | Route drivers to a free bay; fewer double-parked stops and tickets |
| Retail streets and business districts | Shared loading bays for shop suppliers, with evidence for more or better-placed bays |
| Parking enforcement | Find overstays in loading bays without patrols or cameras |
| Hospitals, campuses and industrial sites | Manage private loading bays and short-stay drop-off areas |
| Research and education | Open, reproducible curb-use data for transport studies |
| Country or region | Why it matters there |
|---|---|
| United States | In downtown Seattle, 41 % of commercial vehicles parked in unauthorized locations (Urban Freight Lab, 2019); cities are now comparing curb programs through US DOT SMART grants (Urban Freight Lab) |
| United Kingdom | Van traffic grew 2.3 % to 59.0 billion vehicle miles in the year to September 2024 (Department for Transport) |
| European Union | Data protection by design is a legal duty (Regulation (EU) 2016/679, Article 25), which favors curb sensors that cannot capture personal data at all |
| South Korea | Online sales reached 25.9 % of retail in 2020, the highest share UNCTAD reported (UNCTAD, 2021), so dense cities carry heavy parcel traffic |
| India | National guidelines for City Logistics Plans ask cities to optimize loading and unloading lots and to track the daily use of each bay and the number of deliveries it serves (DPIIT, Guidelines for Preparing City Logistics Plan); a low-cost open sensor can supply that count without cameras |
| Brazil | São Paulo's traffic agency lists 1,749 on-street truck bays reserved for loading and unloading, paid by digital card for one or two hours (CET São Paulo); sensing could show drivers which are free and show the city how long each stop lasts |
The starting point was San Francisco's SFpark pilot, which set magnetometer "pucks" into metered parking spaces to publish real-time availability. The agency's own sensor data guide records where that approach fell short: the in-ground sensors, 4 in (about 100 mm) across, had batteries intended to last about five years, cut to about three by noise-filtering software, yet some began to fail in late 2012 and early 2013, about a year earlier than expected, and users were advised to aggregate the data by hour and by block to reduce the effect of sensor error (SFMTA, Parking Sensor Data Guide, 2013). LoadZone takes the same sensing principle to the loading bay, where stops last minutes rather than hours, and answers those lessons in the open: a surface-bonded puck that needs no coring, a cell budget derated 40 % against the five-year target, and firmware and data that a city can inspect rather than rent.
Delivery vehicles double-park when loading bays are occupied or unknown, blocking lanes and bike paths. Drivers cannot see whether a bay is free before they reach it, and cities have no routine record of how their loading bays are used. See the problem statement.
A curbside loading bay occupancy sensor for streets that are not plowed: it shows delivery drivers which bays are free and gives cities data on curb use.
Full design precis: docs/02-concept.md
- Surface-bonded bay sensor puck on a two-part road-marker epoxy bed, 150 mm diameter and 31 mm high, cast polyurethane over potted electronics, one per vehicle slot of about 7 m
- 3-axis magnetometer (LIS2MDL class) for vehicle detection
- FieldNode radio core (STM32WL-class LoRaWAN module) with an internal antenna
- Two AA-size lithium thionyl chloride cells, 7.7 years of life at SF9 (calculated, derated 40 %)
- Open server and data feed that maps onto the Curb Data Specification
- Solar signage option: 7.5 in e-paper sign on an existing pole, powered by a host FieldNode, legible by day; needs a private network server such as TwinKit
TRL 3 calculations (LDZ-CAL-001): state reported in 17.7 s typically and 18.2 s at worst on US915, 7.7 years on one set of cells, and an estimated USD 116 in parts for a two-slot bay against a value-engineering target of USD 120 (USD 4 under it); the sign option (USD 101 plus a USD 139.50 FieldNode) is costed separately. No requirement is now not met. A puck under a parked van reaches a gateway about 340 m away in a street canyon, so each bay needs a gateway within 300 m (R4), and the e-paper sign is a daylight-only aid, unlit at night (R11). Detection of high-clearance trucks (R1), airtime at slow data rates (R5) and the epoxy road bond under braking (R6) are at risk. See the design precis, requirements and design decisions (LDZ-DDR-001, LDZ-DDR-002, LDZ-DDR-003). The parametric model is cad/src/model.py, with STEP files in cad/step/.
The working bill of materials is in bom/bom.csv.
The prototype build plan (LDZ-BLD-001) shows, in pictures, how to make each of the eleven components and put them together in nine steps; nothing has been built yet. Each puck is a polyurethane dome cast in a home-made silicone mould with a printed core, on a printed base tray that holds the cells, radio board and antenna; the puck is then potted solid through a hole in the tray and bonded to a test slab. The sign option is a bought sign face and display housing on two aluminium channel brackets with band clamps. Writing the plan made the design buildable: the base became a tray with a locating ring, fill and vent holes and holders for every part, the dome got a real casting process, and the sign's solid ring clamps became brackets and bands (LDZ-DDR-003, accepted by Amish on 2026-10-02). Every picture is drawn from the model, which checks that each part touches what it should and clears what it should not.
Street furniture and pole mounts must be installed only with the asset owner's permission, by trained crews, with fall protection and traffic management as local rules require.
Pucks are installed in the roadway: work only under a road permit and traffic management. The pucks contain primary lithium thionyl chloride cells: never recharge, short, crush or heat them. Road-marker adhesives and polyurethane casting resins are hazardous; follow their safety data sheets. The surface puck is not for plowed streets, and a sign's legend must follow the city's sign rules and be approved by its traffic engineer before it goes on a street.
| Folder | Contents |
|---|---|
docs/ |
Problem, concept, requirements, calculations, prototype build plan and design decisions |
cad/src/ |
build123d Python source, the source of truth for all geometry |
cad/step/, cad/stl/ |
Exported models for FreeCAD, other CAD tools and printing |
cad/drawings/ |
2D sketches and dimensioned drawings |
bom/ |
Bill of materials |
electronics/ |
KiCad schematics and PCB layouts |
firmware/ |
Microcontroller code |
media/ |
Renders, perspectives and photos |
build-log/ |
Dated prototyping notes |
Controlled documents follow the portfolio documentation standard. Each carries a document ID (LDZ-PRC-001 for the precis), a version and a revision history. Branded PDFs are built with python .kit/render.py and attached to GitHub Releases when a document is tagged, for example LDZ-PRC-001/v1.0.
Designed by Amish Chadha. See CONTRIBUTORS.md for roles. To cite this design, use CITATION.cff (GitHub shows it as "Cite this repository").
AI assistance (Claude) was used to accelerate concept renders, prototype documentation and first-pass sizing calculations. Design direction and all decisions are Amish Chadha's, recorded in this repository's decision records (docs/decisions/).
- Hardware (CAD, drawings, BOM, electronics): CERN-OHL-S v2
- Software (firmware, scripts, notebooks): MIT
A project of the Design Molecule lab.