**7. Related Work**

To the best of our knowledge general-purpose architecture aiming to integrate and augmen<sup>t</sup> different tourist data-sources and services, has already been proposed, however neither as a proposal in literature nor as a business product. However, some work in the literature has formerly proposed infrastructures to support the development of smart tourism services.

RADON is an EU-funded project aiming to develop a model-driven DevOps framework capable of managing the entire lifecycle of services developed and deployed on top of FaaS platforms. The scope of this project also includes the porting of ADAMO, an application that combines tourist preferences with a city mobility network and points of interest in order to generate customized routes [41].

In [42], authors describe a model architecture for smart tourism systems (STS) tailored for cultural heritage and territorial data. The proposed architecture prompts a rethinking of the key paradigms: interactive travel, tourist gaze, hospitality, authenticity, and social networking data owing to the exploitation of software as a service platform tailored on the development of ST services.

However, the proposed platforms, while providing valid support for the development of new services, do not address the problem of heterogeneity of source in data and services nor propose an integration pattern to integrate with existing tourism offer.

In [43], the authors proposed a suite of small applications in tourism, by using a recommendation approach and supported in a microservice pattern, via a set of independent deployable services. The authors define the following functions: (1) to sugges<sup>t</sup> routes and point of interest to users with respect to the choice of tourism activities; (2) the mutual supporting tourist and tourism services suppliers with location services; (3) to help communities in the preservation and valorization of cultural heritages; (4) to enable tourist to share their travel experiences to help other travelers in their decision-making process.

The proposed architecture of APERTO5.0 instead does not focus only on single tourism services or vertical tourism experiences but aims to integrate and involve as many data sources as possible to create a richer, more integrated, and unified tourism experience.

### **8. Discussion and Conclusions**

In conclusion, the novel development opportunity of smart tourism via a pervasive redesign due to modern IT technologies constitutes a unique occasion toward more sustainable, inclusive, and culturally rich tourism tailored according to customer preferences. In this context, the cooperation with all stakeholders of the territory and the stimulation of local action of social and sensor sensing are necessary to consider and provide an unprecedented big data amount to enable and stimulate even more the creation of smarter and more connected tourism services. However, the deep fragmentation of services and technologies adopted by different actors in tourism that characterize also the whole information provided by customer sensing and IoTs heterogeneity deeply clash with an effective organization of smart tourism.

It is the authors' opinion that the introduction of APERTO5.0 can provide a significant contribution to Academia and Tourism business development and management. This paper has proposed APERTO5.0 as an architecture aiming to address the problem of heterogeneity by providing a unifying view in which any tourist item (data, service, and agents) can become part of an integration mosaic capable of accommodating any new possible element. Such an unifying environment is the authors' main design goal and represents a major innovation in academic research for its novel model and the innovative technological solutions employed in its first proposed prototype.

In fact, the introduction of APERTO5.0 as support to the structuration of ST services over the "Francigena way" allowed us to underline the grea<sup>t</sup> adaptability in providing a unifying support over the huge amount of information available. In this context, this solution introduces an innovative mechanism to query and gather data coming from complex scenarios in an efficient and scalable way enabling the introduction of actions of social sensing and partner involvement. A FaaS layer handles gathered data, to cope with the variety and availability fluctuations of information and then processed with Apache Spark. Results showed that the proposed platform is able to collect and process information in parallel, also traversing complex infrastructure topologies, with a resulting end-to-end latency lower than 1 s.

Considering the contribution of APERTO5.0 to the tourism managemen<sup>t</sup> field, owing to the cooperation with realities of the territory and the explored use-case of tourism paths, authors already demonstrated the effectiveness of proposed solutions. However, a better exploration of the potentiality of a pervasive application of concepts and possibilities opened by Aperto5.0 in the many possible tourism facets and in particular in tourism managemen<sup>t</sup> deserves better exploration by field experts.

As future directions, authors will work both in the use case of pilgrim's paths and in new technical widening directions.

Along the first line, through collaboration with tourism realities in the territory, we aim to further develop APERTO5.0 to support novel innovative and integrated forms of tourism. The integration of also other forms and use cases of tourism can increase furthermore the value of the interconnection of services and information toward a unified and pervasive support to ST development.

To follow novel technical directions, we plan to exploit better the potential of serverless infrastructure in providing fast fine-grained scaling of resources and in reducing the timeto-market development of new services. To pursue these capabilities, future research work will focus on optimizations and the introduction of new capabilities in the FaaS platform, so as to easily meet the needs of ST use cases. The main idea is to explore a more pervasive integration of this model of cloud computing with decentralized deployments over the socalled cloud continuum [41] and exploit data and service locality to achieve lower latencies and finer-grained customization of the platform. Moreover, we stress that a fast diffusion of the FaaS is highly likely, so many other features will be available very soon for even better performances.

**Author Contributions:** Conceptualization, A.S. and A.C.; Data curation, A.S., T.V. and A.C.; Formal analysis, A.C.; Funding acquisition, A.C.; Investigation, A.S. and T.V.; Methodology, A.S. and A.C.; Project administration, A.C.; Software, A.S. and T.V.; Supervision, A.S. and A.C.; Validation, T.V. and A.C.; Writing—original draft, A.S., T.V. and A.C. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research received no external funding.

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

**Informed Consent Statement:** Not applicable.

**Data Availability Statement:** Data available on request due to restrictions. **Acknowledgments:** We would like to thank Stefano Maggiore and Imola Informatica for their collaboration in the definition and refinement of the Francigena way use case and for the providing of the computational resources employed during tests.

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