Recently itinerary recommendation systems have become a trending topic. Many tourist depends on online Location-Based Social Networking to organize a trip. Despite having so many options tourists find it hectic to arrange an itinerary during popular festival seasons due to a lack of knowledge. To deal with the difficulty and to recommend itineraries to tourists during multiple festivals happening simultaneously a technique called FOSTOUR is proposed. In this work, a modified orienteering problem is utilized that incorporates time-based user interest in POIs and popularity of POIs during the festival seasons to recommend the itinerary. The experiment is conducted on a real-time Foursquare dataset. The performance analysis is done regarding precision, recall, f1-score, and accuracy. A few baselines are also introduced to compare the performances of FOSTOUR with them. The results of the comparison clearly show that the performance of FOSTOUR outperforms the rest of the baseline algorithms and recommends a suitable itinerary to tourists during multiple festivals occurring at a time.