top of page
grad-purple-blue.png

When sensors fail: simple, mixed ways to collect data

Methodology

Training

Intervention

Tool

Data

Study

Use case

THINGS, based on the ELABORATOR Lighthouse city reports

Overview

The lesson in one sentence: combine digital measurement with simple fallbacks, because sensors in public space often take longer or deliver less than planned.

The ELABORATOR Lighthouse cities used surveys, workshops, manual counts, video, traffic models and sensors. Some sensor projects gave valuable insights; others were harder than expected. The cities concluded that evaluation should be "robust but flexible", combining digital methods with manual observation and input from stakeholders.


Highlights

Zaragoza: manual counts saved the baseline. Radar sensors around schools needed several permits at once and expensive installation. The data could not be used, and there was no budget to try again. Manual counts in June and October 2025 replaced them.

Copenhagen: cheap counters, used for trends. Telraam counters protect privacy by processing video on the device, so single counts cannot be checked. Some counted cargo bikes as trucks. Averaged over months, they still show long-term changes.

Helsinki: people and e-scooters as sensors. A map survey of 1,778 residents found risky places, a quarter of which were not in the accident records. Sensors on 40 shared e-scooters showed where riders braked or swerved suddenly. Using shared vehicles as data collectors was cost-effective.

Helsinki: testing the tools. At a tram crossing, a Telraam counter worked well. Drone footage was tested but could not yet detect pedestrians accurately enough.

Milan: a sample, not every site. With many small crossing changes, Milan monitored a row of intersections on one street as a representative sample.

Trikala: surveys that reach people. A survey promoted through social media, the city website and QR-code posters on site collected 143 answers in two weeks. On the cycle lanes, 22 sensors alert the police to illegal parking.


Practical relevance

Do

  • Start permits early for any sensors in public space (Zaragoza).

  • Consider cheaper options that need fewer permits, such as portable cameras or manual counts (Zaragoza).

  • Plan a backup data collection from the start (Zaragoza).

  • Treat data collection as continuous, with regular updates and user involvement (Milan).

  • Store only aggregated counts when using data about people's movements, to protect privacy (Copenhagen).

Avoid

  • Spending the whole budget on one sensor campaign with no money to repeat it (Zaragoza).

  • Reading single counts from low-cost counters; use them for trends (Copenhagen).

  • Assuming a new tool is ready: drone footage could not yet detect pedestrians reliably (Helsinki).


Publication details

Article by THINGS, based on the Zaragoza, Copenhagen, Helsinki, Milan and Trikala chapters of ELABORATOR deliverable D5.2, Lighthouse Cities Implementation Final Report (Work Package 5, July 2026), edited by Merja Penttinen, Anu Tuominen and Johannes Mesimäki (VTT). Dissemination level: public.


Related


Disclaimer: The content of this article may be updated or changed, and this page will be updated if it does.

Full materials

Share this page:

Stay in touch!

Sign up to our mailing list to receive exclusive access to our newsletter and stay up to date with the latest project news and events.

Thanks for submitting!

Elaborator logo
  • LinkedIn
  • X

© 2023 Elaborator

Crafted with ♥️ by Threerooms

Co-funded EU logo

Funded by the European Union. Views and opinions expressed are however those of the author(s) only and do not necessarily reflect those of the European Union or CINEA. Neither the European Union nor the granting authority can be held responsible for them.

Civitas-Logo
EU-Mission-Logo

ELABORATOR is part of the CIVITAS Initiative and contributes to the goals of the EU Mission: Climate-Neutral and Smart Cities

UKRI logo [W].png

UKRI supports UK participants with grant numbers 10078745 (iRAP ) and 10069203 (UBRIS)

bottom of page