Model Rekomendasi Latihan Muscle Building Pemula Menggunakan Forward Chaining Berbasis Karakteristik Pengguna

Pasadena Saka (Institut Teknologi dan Bisnis Asia Malang, Kota Malang)
Vivi Aida Fitria (Institut Teknologi dan Bisnis Asia Malang, Kota Malang)

TIN: TERAPAN INFORMATIKA NUSANTARA
Vol 7 No 1 (2026): June 2026

Beginners in muscle building training often have difficulty determining an exercise program that suits their body condition, training goals, available time, training location, equipment availability, and injury history. This study aims to design and develop a muscle building exercise recommendation system for beginners using the forward chaining method. The input data include injury history, height and weight, resistance training experience, basic technique mastery, previous training frequency, training goal, training focus, training location, equipment availability, number of training days, and training duration per session. Height and weight are used to calculate Body Mass Index (BMI), while training experience, basic technique mastery, and previous training frequency are used to determine the beginner level. The knowledge base is arranged into six rule groups: injury rules, body condition rules, beginner level rules, initial intensity rules, program adjustment rules, and exercise recommendation rules. The inference process matches user facts with the rule set step by step until an exercise recommendation is produced. The system output includes an exercise program, schedule, duration, intensity, exercise examples, body condition notes, injury notes, and matched rule information. Initial functional testing and knowledge base validation were conducted using 30 user input combination scenarios. The results showed that 26 scenarios were suitable and 4 scenarios were not yet suitable, resulting in a suitability percentage of 86.67%. These findings indicate that the system can run most of the inference flow in the tested scenarios, but additional rules are still needed for uncovered fact combinations.