The exhaust gas regeneration (EGR) also called exhaust gasre-circulation system in an engine of construction machine (CM) oftengets clogged due to various ways of driving the machines. Currently, theredoes not exist any model that can predict clogging for maintenance planning.Hence, clogging is only recognized when it has occurred, and oftencauses the CM to drop out. Engines still operated despite clogging causesfrequent cold engine running, and excessive exhaustion of nitrogen, whichleads to loss of the engine's performance and reduces their lives.We propose an approach that builds on virtual key sensors. Virtual keysensors are usually used to replace real sensors. However, we proposeto compare the virtual and the real sensor outcomes. If di erences betweenthe estimated and the real value emerge, we assume changes ofthe systems because of, e.g., clogging or leakage in pipes. EGR pressureis identi ed as an important sensor to estimate clogging. A virtualsensor of EGR pressure is built from other real sensors based on a polynomialregression model [1]. The error between the real and the virtualEGR pressure sensors varies between 5-10% depending on the driver'sbehaviors. The model discriminates the ideal ways of working and abnormalities.Moreover, we suggest to adapt the weights of the regressionmodel to other engine types of the same engine family based on onlinestochastic gradient descent algorithm. Since the deployed regression andadaptation algorithms are computationally inexpensive, the approachcould be applied using existing CM micro controllers.