In learning systems and environment research, intelligent tutoring and personalisation are considered the two most important factors. An Intelligent Tutoring System can serve as an effective tool to improve problem-solving skills by simulating a human tutor’s actions in implementing one-to-one adaptive and personalised teaching. Thus, in this research, a solution-based intelligent tutoring system (SITS) is proposed. It benefits from Bayesian networks in managing uncertainty based on the probability theory for the process of decision-making so as to aid students learn computer programming. Additionally, SITS benefits from a multi-agent system that employs an automatic text-to-flowchart conversion approach to engage novice programmers in flowchart development with the aim of improving their problem-solving skills. Finally, the performance of SITS is investigated through an experimental study. It is revealed that SITS is not only capable of boosting students’ learning interest, attitude and technology acceptance, but it also helps students achieve more in terms of problem-solving activities.