Heavy duty machine operators rely on call center service technicians for support in the field. From general inquiries to advanced troubleshooting of problems, this expert support is critical to maintain consistent equipment performance for day-to-day operations. Since any downtime can potentially delay work and lead to increased project time and costs, the aim is to provide customers with rapid and proficient technical support to minimize these adverse events as quickly as possible. In this work, we show how interactive text mining can systematically discover the common problems as well as the real-world verifiable solutions found in unstructured text fields entered by the service technicians during each call. Through frequent phrases and term clustering, we can refine the expert knowledgebase to provide a more standard probabilistic approach to troubleshooting reoccurring issues. Furthermore, this newfound information can be proactively utilized to potentially preempt certain types of incoming calls with more pertinent information available to operators, as well as guide future engineering efforts to mitigate common machine problems in the first place.