Knowledge-Based Decision Support System for Hotel Room Recommendation and Maintenance Prioritization Using Mamdani Fuzzy Logic

Abstract

To improve service quality, hotel management must make decisions quickly and accurately, particularly in providing hotel room recommendations to guests and determining room maintenance priorities by reception staff. However, these decisions are often challenged by uncertainty and subjective considerations. This research is primarily driven by the goal to construct a knowledge-based decision support system. at Hotel Pusaka Mulya that not only facilitates hotel administrative management but also supports accurate room recommendation decisions based on guest preferences and effective prioritization of room maintenance activities. A knowledge-based decision support system leveraging the Mamdani fuzzy reasoning technique is developed as the core of this proposed framework.. The input variables for room recommendation include price, facilities, comfort level, and number of occupants, while the output variable is the room type, consisting of Standard , Standard 1 , Superior 1 , Superior 2 , and Superior 3 . Meanwhile, the input variables for maintenance priority determination are cost, time, and level of damage, with output categories classified as low, medium, and high priority. The Mamdani fuzzy inference process consists of four stages: fuzzification, implication, aggregation using the MAX operator, and defuzzification using the centroid method. The software development process follows the waterfall model, encompassing the phases of analysis, design, implementation, and testing. The testing results demonstrate that the Mamdani Fuzzy  is capable of generating recommendations efficiently and accurately in accordance with the defined decision criteria based on an 80% validation accuracy for room recommendations and an 85% accuracy for determining improvement priorities.