Basic chaining techniques for rule base systems are forward and backward chaining. However, many rule engines employ hybrid methods to involve both chaining techniques to provide more advantages comparatively with those utilizing a single technique. This has proven to have a significant increase on the performance and accuracy of the rule engine. An engine employing pure forward chaining techniques are known at times to generate many facts and hence divert from the original task. In the same manner, pure backward chaining techniques are known to be set on a single conclusion. Sometimes it might not come up with the desired results due to incomplete or unsupportive facts. Given a changeable knowledge base that is continuously being updated with new facts, there exists a need for an engine that adapts to such conditions whilst being optimized not to degrade with time. This study explores the chaining techniques that are being incorporated and uses hybrid chaining in rule based systems to perhaps help calibrate and propose a model for an engine to perform at its maximum potential whilst its constantly being updated. © 2016 American Scientific Publishers. All rights reserved.