Predictive maintenance of equipment

A method for predictive maintenance of an equipment, the method comprising:

• receiving, by a first processor of an edge computing node of a plurality of edge computing nodes, a raw input data from a respective data capturer of a plurality of data capturers, wherein the raw input data pertains to at least one variable corresponding to a condition of an equipment;
• processing, by the first processor, the raw input data to obtain a representative data, wherein the representative data corresponds to a representative set of nodes selected from the plurality of edge computing nodes, and comprises:
• an insight pertaining to a deviation in the at least one variable, and
• a corresponding remedial action to be taken to correct the deviation,
• wherein the deviation is related to a deterioration in the condition of the equipment;
• recommending, by the first processor, the remedial action in the form of a feedback;
• and
• performing, by an actuator coupled with the equipment, an automated actuation based on the feedback,
• wherein the plurality of edge computing nodes are segmented into multiple groups, and wherein the representative set of nodes are selected from each of the multiple groups.

A non-transitory computer readable medium comprising machine executable instructions that are executable by a processor to:

• receive a raw input data from a respective data capturer of a plurality of data capturers, wherein the raw input data pertains to at least one variable corresponding to a condition of an equipment;
• process the raw input data to obtain a representative data, wherein the representative data comprises an insight pertaining to a deviation in the at least one variable and a corresponding remedial action to be taken to correct the deviation, wherein the deviation is related to a deterioration in the condition of the equipment;
• perform an automated actuation based on the corresponding remedial action;
• receive a selective dataset from a representative set of nodes, wherein the selective dataset comprises the representative data from the representative set of nodes selected from a plurality of edge computing nodes, wherein the plurality of edge computing nodes are segmented into multiple groups such that the representative set of nodes are selected from each of the multiple groups;
• and
• train, based on the selective dataset, an artificial intelligence (AI) model, through a collaborative learning for automated prediction of failure state patterns pertaining to the condition of the equipment.