Training inventory management robots using digital twins, trained machine learning models, and human feedback
A VCN process may receive information associated with a value chain network. A VCN process may provide the information to a set of Artificial Intelligence (AI)-based learning models, wherein at least one member of the set of AI-based learning models is trained to classify at least one of: an operating state, a fault condition, an operating flow, or a behavior of the value chain network and at least one member of the set of AI-based learning models is trained on the training data set to determine, upon receiving the classification of the at least one of: the operating state, the fault condition, the operating flow, or the behavior, a task to be completed for the value chain network. A VCN process may configure a robotic process automation system to execute the task to facilitate an improvement in the value chain network.
1 . A computer-implemented method comprising:
receiving, by a computing device, information associated with a value chain network, wherein:
the information includes observations of a human performing a set of tasks within the value chain network, and
the information is generated by at least one of: a set of sensors of a set of value chain network entities, a set of IoT devices configured to collect data relating to the set of value chain network entities, or a set of APIs configured to publish data relating to the set of value chain network entities;
providing, by the computing device, the information to a set of machine learning models, wherein:
the set of machine learning models includes a modular neural network,
the modular neural network includes a set of independent neural networks moderated by an intermediary, and
each neural network of the set of independent neural networks is associated with a respective value chain entity of the set of value chain network entities;
training, by the computing device, each machine learning model of the set of machine learning models on a training data set to classify at least one of: an operating state, a fault condition, an operating flow, or a behavior of the value chain network, wherein the training data set includes:
the information associated with the value chain network, including the observations of the human performing the set of tasks within the value chain network,
digital twin simulations of the human performing the set of tasks within the value chain network, and
operating data associated with the set of value chain network entities;
automatically generating, by the computing device executing the trained set of machine learning models, a task traditionally performed by a human to be completed for the value chain network from the set of tasks based on a classification generated by the trained set of machine learning models, by:
receiving, by the intermediary, inputs from the set of independent neural networks, and
aggregating, by the intermediary, the inputs to generate the task;
configuring, by the computing device executing the trained set of machine learning models, a robotic process automation system to train a set of physical robots to execute the task to facilitate an improvement in the value chain network;
training, by the computing device executing the robotic process automation system, at least one robot of the set of physical robots to execute the task;
instructing, by the computing device, the at least one robot of the set of physical robots to execute the task;
gathering, by the at least one robot of the set of physical robots, feedback on an outcome of the executed task;
providing, by the at least one robot of the set of physical robots, the feedback to the computing device; and
retraining, by the computing device, at least one trained member of the trained set of machine learning models on the feedback.
2 . The computer-implemented method of claim 1 , wherein the task includes automatically processing one or more orders for an item in the value chain network based upon, at least in part, the information.
3 . The computer-implemented method of claim 2 , wherein the automatically processing the one or more orders for the item in the value chain network includes:
extracting order data of the one or more orders from one or more sources; and
automatically inputting the order data extracted from the one or more sources into an order management system.
4 . The computer-implemented method of claim 1 , wherein the task includes:
monitoring an inventory level for an item in the value chain network; and
automatically generating one or more purchase orders for the item when the inventory level for the item in the value chain network falls below a threshold.
5 . The computer-implemented method of claim 4 further comprising:
tracking a shipment of the item in real-time; and
automatically updating the inventory level for the item in the value chain network based upon, at least in part, tracking the shipment of the item in real-time.
6 . The computer-implemented method of claim 1 , wherein the task includes:
extracting invoice data from one or more invoices from one or more sources; and
automatically inputting the invoice data extracted from the one or more sources into an accounting system.
7 . The computer-implemented method of claim 1 , wherein the set of value chain network entities includes at least one of: products, suppliers, producers, manufacturers, retailers, businesses, owners, operators, operating facilities, customers, consumers, workers, mobile devices, wearable devices, distributors, resellers, supply chain infrastructure facilities, supply chain processes, logistics processes, reverse logistics processes, demand prediction processes, demand management processes, demand aggregation processes, machines, ships, barges, warehouses, maritime ports, airports, airways, waterways, roadways, railways, bridges, tunnels, online retailers, ecommerce sites, demand factors, supply factors, delivery systems, floating assets, points of origin, points of destination, points of storage, points of use, networks, information technology systems, software platforms, distribution centers, fulfillment centers, containers, container handling facilities, customs, export control, border control, drones, robots, robotic handling systems, 3D printers, vehicles, autonomous vehicles, hauling facilities, waterways, or port infrastructure facilities.
8 . The computer-implemented method of claim 1 , wherein the trained set of machine learning models includes at least one of: a transformer model, a convolutional neural network, a deep learning model trained on a set of outcomes of a value chain network entity, a supervised model, a semi-supervised model, an unsupervised model, or a reinforcement model.
9 . The computer-implemented method of claim 1 , wherein the training data set for the trained set of machine learning models includes at least one of: a set of objects or events that are labeled to classify the set of objects or events according to a classification taxonomy that includes at least one of: the operating state, the fault condition, the operating flow, or the behavior.
10 . The computer-implemented method of claim 1 , wherein:
the at least one robot of the set of physical robots includes a dynamic vision system configured to gather at least one of images or videos of a target object associated with the task to identify the target object; and
the feedback includes the at least one of images or videos of the target object captured by the dynamic vision system while the at least one robot executes the task.
11 . The computer-implemented method of claim 10 , further comprising updating, by the computing device executing the at least one retrained member of the set of machine learning models, at least one optical parameter of the dynamic vision system.
12 . The computer-implemented method of claim 10 , wherein the observations of the human performing the set of tasks within the value chain network include observations of the human interacting with the target object.
13 . The computer-implemented method of claim 1 , wherein the execution of the task includes performing a maintenance activity on at least one value chain network entity of the set of value chain network entities.
14 . The computer-implemented method of claim 1 , wherein the retraining the at least one trained member of the set of machine learning models includes reconfiguring a set of weights of the at least one trained member of the set of machine learning models.
15 . A computing system including one or more processors and one or more memories configured to perform operations comprising:
receiving, by a computing device, information associated with a value chain network, wherein:
the information includes observations of a human performing a set of tasks within the value chain network, and
the information is generated by at least one of: a set of sensors of a set of value chain network entities, a set of IoT devices configured to collect data relating to the set of value chain network entities, or a set of APIs configured to publish data relating to the set of value chain network entities;
providing, by the computing device, the information to a set of machine learning models, wherein:
the set of machine learning models includes a modular neural network,
the modular neural network includes a set of independent neural networks moderated by an intermediary, and
each neural network of the set of independent neural networks is associated with a respective value chain entity of the set of value chain network entities;
training, by the computing device, each machine learning model of the set of machine learning models on a training data set to classify at least one of: an operating state, a fault condition, an operating flow, or a behavior of the value chain network, wherein the training data set includes:
the information associated with the value chain network, including the observations of the human performing the set of tasks within the value chain network,
digital twin simulations of the human performing the set of tasks within the value chain network, and
operating data associated with the set of value chain network entities;
automatically generating, by the computing device executing the trained set of machine learning models, a task traditionally performed by a human to be completed for the value chain network from the set of tasks based on a classification generated by the trained set of machine learning models, by:
receiving, by the intermediary, inputs from the set of independent neural networks, and
aggregating, by the intermediary, the inputs to generate the task;
configuring, by the computing device executing the trained set of machine learning models, a robotic process automation system to train a set of physical robots to execute the task to facilitate an improvement in the value chain network;
training, by the computing device executing the robotic process automation system, at least one robot of the set of physical robots to execute the task;
instructing, by the computing device, the at least one robot of the set of physical robots, to execute the task;
gathering, by the at least one robot of the set of physical robots, feedback on an outcome of the executed task;
providing, by the at least one robot of the set of physical robots, the feedback to the computing device; and
retraining, by the computing device, at least one trained member of the trained set of machine learning models on the feedback.
16 . The computing system of claim 15 , wherein the task includes automatically processing one or more orders for an item in the value chain network based upon, at least in part, the information.
17 . The computing system of claim 16 , wherein the automatically processing the one or more orders for the item in the value chain network includes:
extracting order data of the one or more orders from one or more sources; and
automatically inputting the order data extracted from the one or more sources into an order management system.
18 . The computing system of claim 15 , wherein the task includes:
monitoring an inventory level for an item in the value chain network; and
automatically generating one or more purchase orders for the item when the inventory level for the item in the value chain network falls below a threshold.
19 . The computing system of claim 18 , wherein the operations further comprise:
tracking a shipment of the item in real-time; and
automatically updating the inventory level for the item in the value chain network based upon, at least in part, tracking the shipment of the item in real-time.
20 . The computing system of claim 15 , wherein the task includes:
extracting invoice data from one or more invoices from one or more sources; and
automatically inputting the invoice data extracted from the one or more sources into an accounting system.
21 . The computing system of claim 15 , wherein the set of value chain network entities includes at least one of: products, suppliers, producers, manufacturers, retailers, businesses, owners, operators, operating facilities, customers, consumers, workers, mobile devices, wearable devices, distributors, resellers, supply chain infrastructure facilities, supply chain processes, logistics processes, reverse logistics processes, demand prediction processes, demand management processes, demand aggregation processes, machines, ships, barges, warehouses, maritime ports, airports, airways, waterways, roadways, railways, bridges, tunnels, online retailers, ecommerce sites, demand factors, supply factors, delivery systems, floating assets, points of origin, points of destination, points of storage, points of use, networks, information technology systems, software platforms, distribution centers, fulfillment centers, containers, container handling facilities, customs, export control, border control, drones, robots, robotic handling systems, 3D printers, vehicles, autonomous vehicles, hauling facilities, waterways, or port infrastructure facilities.
22 . The computing system of claim 15 , wherein the trained set of machine learning models includes at least one of: a transformer model, a convolutional neural network, a deep learning model trained on a set of outcomes of a value chain network entity, a supervised model, a semi-supervised model, an unsupervised model, or a reinforcement model.
23 . The computing system of claim 15 , wherein the training data set for the trained set of machine learning models includes at least one of: a set of objects or events that are labeled to classify the set of objects or events according to a classification taxonomy that includes at least one of: the operating state, the fault condition, the operating flow, or the behavior.
24 . The computer-implemented method of claim 10 , wherein the training data set includes digital twin simulations of the dynamic vision system.
25 . The computing system of claim 15 , wherein:
the at least one robot of the set of physical robots includes a dynamic vision system configured to gather at least one of images or videos of a target object associated with the task to identify the target object; and
the feedback includes the at least one of images or videos of the target object captured by the dynamic vision system while the at least one robot executes the task.
26 . The computing system of claim 25 , wherein the computing device is configured to update, by executing the at least one retrained machine learning model, at least one optical parameter of the dynamic vision system.
27 . The computing system of claim 25 , wherein the observations of the human performing the set of tasks within the value chain network include observations of the human interacting with the target object.
28 . The computing system of claim 25 , wherein the training data set includes digital twin simulations of the dynamic vision system.