Demand-responsive raw material management system

A computerized method for raw material procurement, the computerized method comprising:

• calculating an expected demand for a product at a future point in time;
• identifying at least one of:
• an environmental condition or an environmental event;
• estimating a raw material availability of a raw material at the future point in time based on the expected demand and the at least one of:
• the environmental condition or the environmental event;
• calculating a required raw material amount of the raw material to manufacture the product at the future point in time based on the expected demand and on the at least one of:
• the environmental condition or the environmental event;
• autonomously configuring a futures contract for procurement of at least a portion of the required raw material amount in response to the required raw material amount calculation exceeding the raw material availability;
• retrieving a pre-determined risk tolerance of an entity that procures the raw material;
• generating a risk threshold based on the pre-determined risk tolerance of the entity that procures the raw material,
• wherein the pre-determined risk tolerance indicates a difference between a baseline cost of the raw material and a futures cost of the raw material,
• wherein the autonomously configuring the futures contract includes providing, by a robotic process automation (RPA) service, a set of instructions for autonomously configuring the futures contract,
• wherein the RPA service trains a neural network on a training data set to provide the set of instructions for autonomously configuring the futures contract,
• wherein the training data set includes at least one of:
• historical data, feedback from outcomes, or human interactions involved in contract negotiations,
• wherein the providing the set of instructions, by the RPA service, includes performing a set of de-risking algorithms to configure a set of terms and conditions for the futures contract,
• wherein the autonomously configuring the futures contract is based at least in part on the risk threshold and the set of instructions from the RPA service,
• wherein the autonomously configuring the futures contract includes executing a smart contract for the futures contract, and
• wherein the smart contract is implemented on a blockchain;
• executing the smart contract in response to a risk exceeding the risk threshold,
• wherein the risk is associated with a possibility of the futures cost of the raw material exceeding a maximum price the entity that procures the raw material is willing to pay, and
• wherein the RPA service includes an interface for an entity associated with the executed smart contract to provide feedback on the executed smart contract;
• and
• retraining the neural network based on (i) an outcome of the executed smart contract and (ii) the feedback received from the entity associated with the executed smart contract,
• wherein the retraining of the neural network includes placing gates on the neural network to render it a gated neural network that balances learning with a need to diminish certain inputs.

A computer system comprising:

• memory hardware configured to store computer-executable instructions;
• and
• processor hardware configured to execute the instructions,
• wherein the instructions include:
• calculating an expected demand for a product at a future point in time;
• identifying at least one of:
• an environmental condition or an environmental event;
• estimating a raw material availability of a raw material at the future point in time based on the expected demand and the at least one of:
• the environmental condition or the environmental event;
• calculating a required raw material amount of the raw material to manufacture the product at the future point in time based on the expected demand and on the at least one of:
• the environmental condition or the environmental event;
• autonomously configuring a futures contract for procurement of at least a portion of the required raw material amount in response to the required raw material amount calculation exceeding the raw material availability;
• retrieving a pre-determined risk tolerance of an entity that procures the raw material;
• generating a risk threshold based on the pre-determined risk tolerance of the entity that procures the raw material,
• wherein the pre-determined risk tolerance indicates a difference between a baseline cost of the raw material and a futures cost of the raw material,
• wherein the autonomously configuring the futures contract includes providing, by a robotic process automation (RPA) service, a set of instructions for autonomously configuring the futures contract,
• wherein the RPA service trains a neural network on a training data set to provide the set of instructions for autonomously configuring the futures contract,
• wherein the training data set includes at least one of:
• historical data, feedback from outcomes, or human interactions involved in contract negotiations,
• wherein the providing the set of instructions, by the RPA service, includes performing a set of de-risking algorithms to configure a set of terms and conditions for the futures contract,
• wherein the autonomously configuring the futures contract is based at least in part on the risk threshold and the set of instructions from the RPA service,
• wherein the autonomously configuring the futures contract includes executing a smart contract for the futures contract, and
• wherein the smart contract is implemented on a blockchain;
• executing the smart contract in response to a risk exceeding the risk threshold,
• wherein the risk is associated with a possibility of the futures cost of the raw material exceeding a maximum price the entity that procures the raw material is willing to pay, and
• wherein the RPA service includes an interface for an entity associated with the executed smart contract to provide feedback on the executed smart contract;
• and
• retraining the neural network based on (i) an outcome of the executed smart contract and (ii) the feedback received from the entity associated with the executed smart contract,
• wherein the retraining of the neural network includes placing gates on the neural network to render it a gated neural network that balances learning with a need to diminish certain inputs.