Municipal maintenance is one of the least visible and most paper-heavy parts of city work. In Fujisawa, as in many Japanese cities, officers inspecting parks and public spaces wrote their findings on paper, sometimes photographing the problem separately and noting the location by hand. Those records were carried back to the office, transcribed, assigned to a department and followed up manually.
The cost was not only staff time. Every manual handover between spotting a problem and fixing it was a point at which a report could be delayed, misrouted or lost, and the city had no way of detecting which reports had gone missing. Because paper records could not be aggregated, there was also no view of which facilities failed repeatedly, where illegal dumping recurred, or how long repairs actually took.
Citizen reporting applications such as FixMyStreet address a different problem. They widen participation, but they also produce submissions of uneven quality that someone has to triage before any work can begin.
MinaRepo was developed by Jin Nakazawa’s laboratory at Keio University’s Shonan Fujisawa Campus as a participatory sensing application built around civil officers rather than the general public. Officers already cover the city as part of their duties, know what they are looking at, and file reports that are validated by virtue of who filed them — which removes the triage burden that open citizen reporting creates.
In the field, an officer who finds a problem opens the application, selects a category, takes a photograph and submits. Location and time are captured automatically. On the receiving side, a data management system stores and classifies the report, routes it to the department responsible and tracks its status, so the record is digital from the moment of observation rather than from the moment someone types it up.
In Fujisawa the workflow was connected to a team collaboration platform, so a report created a flow of information among the staff who would act on it instead of sitting in a queue. The approach began in the EU–Japan ClouT project and was carried into BigClouT, where MinaRepo ran on the project’s distributed city data platform. The underlying model was published as “Accelerating Urban Science by Crowdsensing with Civil Officers” (UbiComp/ISWC 2018 Adjunct).
Fujisawa, Japan (2016–2019). Deployed with Fujisawa City under the EU–Japan BigClouT project to replace the paper-based park and public realm maintenance workflow. City officers used MinaRepo in the field; reports were routed digitally and connected to the team collaboration platform used by cleaning and maintenance staff. Shown at Keio’s Open Research Forum in 2017 and published at UbiComp/ISWC 2018.
Tsukuba, Japan. The same photo-and-description reporting model was carried into HukuRepo, which targeted foreign visitors rather than city officers and ran on the same BigClouT platform.
Transferability. Because the reporting model is independent of the issue categories, it applies to any municipal service whose staff already move through the city as part of their work — road and footpath inspection, street furniture, drainage, street lighting or vegetation management.
Get in touch to explore deployment opportunities or propose a complementary approach.
