Obesity is defined as abnormal or excessive fat accumulation that presents a risk to health according to the World Health Organization (WHO). Pediatric or childhood obesity is the most prevalent nutritional disorder among children and adolescents worldwide. In pediatric or childhood obesity, constant monitoring of the pediatric patients by health experts is required to provide efficient obesity management and treatment. Therefore, the patients are examined on a regular basis, the measurements are compared against predefined percentile values and the development of the pediatric patient is examined. This study discusses the design, implementation, and potential use of an ontology-based obesity management and consultation system which is a decision support system for health experts during treatments of the children and adolescent patients. The system does not only share instant gathered medical data to health experts but also examines the data as a smart medical assistant. The system includes an ontology-based inference engine module, which is a decision support module, and used to infer certain personalized suggestions for patients. Suggestions in four categories emerged as a result: (1) Development Feedback Suggestions, (2) Calorie Intake Suggestions and Physical Activities, (3) Mom Suggestions, and (4) Obesity Treatment Stage Suggestions. The methodologies applied and main technical contributions are discussed in three aspects: (1) Obesity Tracking Ontology, (2) Semantic Web Rule Knowledge base, and (3) Inference Engine Module. In this study, unlike other similar studies, ontology and rule based smart medical assistant which have different functionalities from adults' obesity management is considered especially for obesity management of children and adolescents. The system also includes intensive pediatric health care expert involvement. Eighty case studies from real anonymous pediatric patients are analyzed and discussed in this experimental study. The results retrieved from 80 case studies are promising in demonstrating the applicability, effectiveness and efficiency of the proposed approach. The inference engine module of the proposed system can be integrated semantically into intelligent and distributed decision support systems, and the system ontology can be used as a knowledge base in similar systems.