Product closing the loop

A method for managing physical assets in a manufacturing system, the method comprising:

• creating a plurality of product controllers, each product controller corresponding to a distinct physical asset or product in the manufacturing system;
• executing the plurality of product controllers, wherein during execution each product controller performs operations comprising:
• varying activities performed by the product controller according to a lifecycle state of the corresponding physical asset or product including design, production and operation states;
• storing a desired product state for a physical asset or product during the design state, wherein the desired product state is determined by setting a set of values based on information from inputs determined using a machine learning model;
• collecting sensor data received from one or more sensors in the manufacturing system,
• determining an actual product state based on the collected sensor data;
• determining one or more actions to be performed on one or more physical assets in the manufacture system to yield the desired product state;
• and
• transmitting control instructions for performing the one or more actions to one or more operation controllers in the manufacturing system and
• executing a plurality of operation controllers on a plurality of control computers, wherein each operation controller performs operations comprising:
• provide control instructions to a physical asset, and
• receive sensor data from sensors associated with the physical asset
• wherein each operation controller is associated to at least one of the product controllers in a cascaded configuration the product controller acting as a master to the operation controller.

A system for managing physical assets in a manufacturing system, the system comprising:

• a plurality of digital twins corresponding to a plurality of physical assets in the manufacturing system;
• a plurality of operation controllers executing within a plurality of control computers, wherein each operation controller controls operation of a machine in the manufacturing system based on process image data stored in a corresponding control computer;
• a plurality of product controllers corresponding to the plurality of digital twins, wherein each product controller is configured to:
• vary its activities according to a lifecycle state of a product associated with the product controller;
• during a design stage, determining a desired product state by setting values determined by inputs identified using a machine learning model;
• in the product controller, collect sensor data from one or more sensors in the manufacturing system,
• in the product controller, determine an actual product state for a physical asset in the manufacturing system,
• in the product controller, determine a desired product state based on the digital twin corresponding to the physical asset, and
• if the actual product state deviates from the desired product state, the product controller providing instructions to one or more of the operation controllers to perform one or more actions on the physical asset.