Smart Poqueira: Predicting Rural Parking Lot Feasibility with Sensor-Questionnaire Integration

Abstract

We introduce a dataset comprising 525 instances of visitor behavior data from Pampaneira, Bubión, and Capileira in the Sierra Nevada National Park, Granada, Spain. Collected in January, March, and July 2023, the questionnaires excluded locals and residents. Conducted in parking lots, the questionnaires gathered information like license plate numbers, residential postcodes, visit frequency, and overnight stays. Additionally, data from four Hikvision license plate recognition (LPR) cameras tracking vehicle movement in each village during the same period supplement the dataset, enhancing understanding of individual mobility patterns. To further enrich the dataset, contextual details such as holiday days, vehicle provenance, and socio-demographic information, aiding in the validation and enhancement of questionnaire-derived data. The dataset comprises 26 variables, including: avg_gross_income: Average gross income of the area of origin of the vehicle. (Float) population: Population size of the city/town of the provenance of the vehicle. (Integer) km_to_POQ: Distance in kilometers between the origin of the vehicle and the destination region (Pampaneira reference point). (Float) avg_nights: Average number of nights the vehicle spent in the area. (Float) std_nights: Standard deviation of the average number of nights. (Float) avg_holiday: Average number of holidays the vehicle spent in the area. (Float) std_holiday: Standard deviation of the average number holidays spent. (Float) avg_workday: Average number of workdays the vehicle spends in the area. (Float) std_workday: Standard deviation of the average number workdays spent. (Float) avg_high_season: Average number of days of high season the vehicle spends in the area. (Float) std_high_season: Standard deviation of the average number of days of high season spent. (Float) avg_low_season: Average number of days of low season the vehicle spends in the area. (Float) std_low_season: Standard deviation of the average number of days of low season spent. (Float) total_distance: Total distance traveled in kilometers by the vehicle within the area. (Float) total_holiday: Number of holidays the vehicle spent in the area. (Integer). total_workday: Number of workdays the vehicle spent in the area. (Integer). total_high_season: Number of high season days the vehicle spent in the area. (Integer). total_low_season: Number of low season days the vehicle spent in the area. (Integer). entry_in_high_season: Total number of entries on high season. (Integer) entry_in_holiday: Total number of entries on holiday. (Integer) nights: Total number of nights. (Integer) fidelity: Number of visits after maintaining fidelity of at least five days. (Float) visits_dif_weeks: Number of different weeks with at least one visit. (Integer) visits_dif_months: Number of different months with at least one visit. (Integer) total_entries: Total number of entries to the zone. (Integer) park_price_will_affect_behaviour: Inclination on the part of the vehicle driver to revisit the area if paid parking were introduced, with 3 response options: "visit more", "visit less" or "does not affect me". This variable serves as a prediction class. (Categorical)