Apparatus and method for processing sensor data to predict future outcomes

A system comprising:

• a set of sensors, each one of the set of sensors generating a set of current values associated with at least one component included on a moving vehicle, the generated set of current values being transmitted over a wireless network;
• and
• an edge computing device that receives the set of current values transmitted over the wireless network, the edge computing device processing the set of current values in real-time using at least one machine learning algorithm to identify a value of a point in time for a failure of one of the at least one component a based on the set of current values and at least one set of past values received, the at least one machine learning algorithm being a dynamic linear regression algorithm utilizing at least one windowing operation, the at least one windowing operation performing a first windowing operation that includes an averaging of the set of current values and the at least one set of past values to produce a current set of first windowed values for the one of the at least one component, and wherein the at least one windowing operation further includes performing a second windowing operation on the current set of first windowed values, along with past sets of first windowed values, to produce a current set of second windowed values that are used to produce results at different points in time by the dynamic linear regression algorithm, the edge computing device further providing, for display to a user, the value of the identified point in time for the failure the at least one component, the value of the identified point in time for the failure displayed in a different color when it is determined that the identified point in time for the failure is less than a threshold value for the point in time for the failure;
• wherein the generated set of current values are transmitted with a low time latency between the generating of the set of current values and the processing of the set of current values.

An apparatus comprising:

• a network interface that receives a set of current values over a wireless network, the set of current values generated from a set of sensors associated with at least one component included on a moving vehicle;
• at least one processing device coupled to the network interface, the at least one processing device processing the set of current values in real-time using at least one machine learning algorithm to identify a value for a value of a point in time for a failure of one of the at least one component a based on the set of current values and at least one set of past values, the at least one machine learning algorithm being a dynamic linear regression algorithm utilizing at least one windowing operation, the at least one windowing operation performing a first windowing operation that includes an averaging of the set of current values and the at least one set of past values to produce a current set of first windowed values for the one of the at least one component, and wherein the at least one windowing operation further includes performing a second windowing operation on the current set of first windowed values, along with past sets of first windowed values, to produce a current set of second windowed values that are used to produce results at different points in time by the dynamic linear regression algorithm;
• a memory coupled to the at least one processing device, the memory configured to store the set of current values and the at least one set of past values;
• and
• an output interface coupled to the at least one processing device, the output device providing, for display to a user, the value of the identified point in time for the failure the at least one component, the value of the identified point in time for the failure displayed in a different color when it is determined that the identified point in time for the failure is less than a threshold value for the point in time for the failure,
• wherein the current set of data values are received over the wireless network with a low time latency between the set of current values being generated and the set of current values being processed.