Desarrollo de sistema de reservaciones de alojamiento y guías locales con asistente virtual eficiente en la parroquia San Andrés

Abstract

In San Andrés, located in the province of Chimborazo, Ecuador, the tourism sector faces problems of inefficiency in the processes of accommodation reservations and difficulties in finding local guides. To address this problem, the present work aimed to develop and implement a system of accommodation and local guide reservations with an efficient virtual assistant. Analytical methods and techniques such as meetings, field observation, and business process model and notation (BPMN) diagrams were used to determine the existing processes. For the development of the application, Python, Django, and Dialogflow technologies were investigated, and the ScrumBan methodology was used, an agile methodology that combines the best of Scrum and Kanban, to guide the development process of the reservation system. This methodology allowed us to organize the work resulting in 11 user stories that cover the following modules: search, reservations, payments, virtual assistant and administration. To evaluate the performance efficiency, the ISO 25010 standard was applied, using the response time as an indicator of temporal behavior, and evaluating the use of resources in RAM and CPU. In this context, a sample of 384 interactions was analyzed, distributed among the different processes of the system. The results obtained showed that the application reached 83.20% efficiency, ranking as Very Good, therefore, the system is efficient in its overall operation.