One embodiment of the present invention includes a method for reactively cleaning a contact mechanism. The method includes measuring contact resistance (CRES) associated with a plurality of electrical contacts of the contact mechanism. The method also includes generating at least one statistic of the measured CRES associated with the plurality of electrical contacts of the contact mechanism, and comparing the at least one statistic of the CRES associated with the plurality of electrical contacts of the contact mechanism with at least one CRES threshold parameter associated with an unacceptable level of CRES. The method further includes cleaning the plurality of electrical contacts of the contact mechanism based on the comparison of the at least one statistic of the CRES and the at least one CRES threshold parameter.