*5.4. Discussion*

Figure 5 shows a comparative view through which the UFM operator can assess the vehicular activity in configurable timeframes and areas. These views fall inside the functionalities provisioned in the second use case, and are all configurable via dedicated high-level graphical interactions. Once the configuration has been set up, a Kibana view is generated and mirrored in the frontend.

Figure 6 shows some sample outputs generated by the UFM scheduler functionality provisioned in the second use case. For this assessment, the algorithm relies on the predictive index and co-locality information identifying nearby interferences, e.g., city events, other public utility maintenance operations announced by the municipality, etc. A maintenance operation has a specific location in space, and a duration in time that could span several days or hours (Figure 6). Indeed, as already discussed, the data are sliced in the time domain: in Figure 6a, the algorithm exploits the daily predictive index to position a maintenance operation in time, while, in Figure 6d, the maintenance operation is, generally, positioned in a series of consecutive (pre-configured) time intervals. The IPPODAMO interface allows an operator to specify this additional search criterion.

In all the charts, the lines denoted in green identify the minimum cost schedules along with potential identified interferences, gathered and reported via the user interface. It is noteworthy to point out that, in the current implementation, interferences do not contribute to the index, but rather serve as additional information guiding the UFM operator to make an informed decision (Figure 6b,c). In addition, the interface allows the UFM operator to customize the weights of the individual parameters, e.g., as shown in Figure 6c, where the only quantity contributing to the index is that derived from the vehicular data.

**Figure 5.** Trend of the vehicular index—scale [0, 100] for better visualization—in the pre-COVID-19, during and post-COVID-19 period. (**a**) Timeline comprising the COVID-19 period. (**b**) Post-COVID-19 period.

**Figure 6.** The UFM scheduler result, providing the operator with potential timeframes (green color) during which to schedule a specific maintenance operation. The final activity index is decomposed in all its constituent values, contributing to the final index. The analysis considers a period of one month, starting from 10 November 2021. (**a**) Equal cost timeframes proposed by the scheduler for a maintenance operation in Via Indipendenza X, Bologna, Italy. (**b**) Identified non-binding interference for a maintenance operation in Via Zamboni, Bologna, Italy, in the interval [10, 23]. (**c**) Equal cost timeframes, vehicular data only, proposed by the scheduler for a maintenance operation in Via Zamboni. (**d**) Scheduling of an urgen<sup>t</sup> intervention in Via Zamboni, relying on the next-day prediction of the activity index.

### **6. Conclusions**

In this work, we presented a Digital Twin solution for the Urban Facility Management process in a smart city context. IPPODAMO is a multi-layer, distributed system making use of a multitude of heterogeneous data sources to accurately depict and predict the dynamics inside a geographical area of interest. The decision support system consists of a wide variety of visualizations, including a scheduler functionality, aiding UFM operators in their maintenance placement activity.

Currently, the solution is being tested in a real operational scenario, and we are studying emergen<sup>t</sup> software behavior, identifying near-to-mid-term directions to extend the software. Of paramount importance is the capability to quantify the benefits of the solution through measurable KPIs. To this end, we are collaborating with the private sector and structuring a qualitative data gathering process that could serve as a basis for the value proposition of the proposal.

**Author Contributions:** Conceptualization, A.B., A.C. (Alessandro Calvio) and L.F.; Data curation, A.B., A.C. (Alessandro Calvio) and A.S.; Investigation, A.B., A.C. (Alessandro Calvio), A.S., L.F. and A.C. (Antonio Corradi); Resources, A.B., L.F. and A.C. (Antonio Corradi); Software, A.B., A.C. (Alessandro Calvio) and A.S.; Validation, A.B., A.C. (Alessandro Calvio) and A.S.; Writing— original draft, A.B., A.C. (Alessandro Calvio) and A.S.; Writing—review and editing, L.F. and A.C. (Antonio Corradi). All authors have read and agreed to the published version of the manuscript.

**Funding:** This work has been partially funded by the Interactive Planning Platform fOr city District Adaptive Maintenance Operations (IPPODAMO) project (CUP: C31J20000000008), a BI-REX Industry 4.0 Competence Center project.

**Institutional Review Board Statement:** Not applicable.

**Informed Consent Statement:** Not applicable.

**Data Availability Statement:** Not applicable.

**Conflicts of Interest:** The authors declare no conflict of interest.
