Detection and correction of robotic process automation failures

A computer-implemented method comprising:

• receiving, by a predictive model, a representation of an execution history of a particular software program, wherein the predictive model has been trained on a corpus of execution histories of software programs in order to be able to estimate root causes of software program failures, wherein persistent storage contains rules related to repairing software programs in a network, and wherein the rules associate indications of software program failures with repair applications implemented by an automated repair controller according to domain knowledge of the network, wherein the repair applications are configured to correct corresponding software program failures;
• generating, by the predictive model and from the execution history, a failure prediction for the particular software program, the failure prediction including an estimated root cause;
• receiving, by an automated repair controller application, the failure prediction from the predictive model;
• based on applying the rules to the failure prediction, determining, by the automated repair controller application, a repair application from the repair applications that is configured to correct the estimated root cause;
• and
• causing, by the automated repair controller application, the repair application to be executed within the network.

An article of manufacture including a non-transitory computer-readable medium, having stored thereon program instructions that, upon execution by a computing system, cause the computing system to perform operations comprising:

• receiving, by a predictive model, a representation of an execution history of a particular software program, wherein the predictive model has been trained on a corpus of execution histories of software programs in order to be able to estimate root causes of software program failures, wherein persistent storage contains rules related to repairing software programs in a network, and wherein the rules associate indications of software program failures with repair applications implemented by an automated repair controller according to domain knowledge of the network, wherein the repair applications are configured to correct corresponding software program failures;
• generating, by the predictive model and from the execution history, a failure prediction for the particular software program, the failure prediction including an estimated root cause;
• receiving, by an automated repair controller application, the failure prediction from the predictive model;
• based on applying the rules to the failure prediction, determining, by the automated repair controller application, a repair application from the repair applications that is configured to correct the estimated root cause;
• and
• causing, by the automated repair controller application, the repair application to be executed within the network.