Fall from height is a perennial problem in the construction industry. Active fall protection system (AFPS) is frequently a must in situations where working conditions are difficult and other controls are not feasible or inadequate. However, the design and selection of AFPS are still problematic in the construction industry. This paper aims to develop an online knowledge-based system, FPSWizard, to support the design and selection of AFPS. The hybrid system adopts a combination of case-based reasoning (CBR) and rule-based reasoning (RBR) to improve retrieval performance. FPSWizard is meant to recommend suitable AFPS based on similar past design cases. Potential end users, such as professional engineers and safety professionals, can use the system as a decision support system when they are selecting and designing a solution to the work-at-height problem at hand. A total of fifty stored cases were created based on actual work scenarios and AFPS designs in the construction industry. A case structure was also created using the AFPS-Ontology. The system was assessed using a leave-oneout cross validation approach, where fifty cases in the case base were used to test the retrieval performance of the system. The hybrid CBR-RBR approach had an average positive predictive value (PPV) (or precision) of 90%. In comparison, a pure CBR approach had an average PPV of 76%. FPSWizard forms an important part of an intelligent system which provides comprehensive solutions to fall from height. This paper also made important strides towards intelligent safety engineering and management in the construction industry.