Microsoft Technology Licensing, LLC

Peeyush Kumar

Ranveer Chandra

Vaishnavi Nattar Ranganathan

Michael McNab Bassani

Vishal Jain

Shivkumar Kalyanaraman

Chetan Bansal

Weiwei Yang

Michael Grant

Srinivasan Iyengar

Riyaz Pishori

Lillian Jane Ratliff

Hui Qing Li

Upendra Singh

Tanuja Hrishikesh Ganu

Jeremy Randall Reynolds

Emmanuel Azuh Mensah

Lucien Werner

Dang Khoa Tran

Upinder Kaur

A computing system including a processor configured to receive training data including, for each of a plurality of training timesteps, training forecast states associated with respective training-phase agents included in a training supply chain graph. The processor may train a reinforcement learning simulation of the training supply chain graph using the training data via policy gradient reinforcement learning. At each training timestep, the training forecast states may be shared between the training-phase agents during training. The processor may receive runtime forecast states associated with respective runtime agents included in a runtime supply chain graph. For a runtime agent, at the trained reinforcement learning simulation, the processor may generate a respective runtime action output associated with a corresponding runtime forecast state of the runtime agent based at least in part on the runtime forecast states. The processor may output the runtime action output.

Reinforcement learning simulation of supply chain graph

The techniques disclosed herein enable systems to enable multi-market optimization of renewable energies using data-driven models. To achieve this, a model retrieves a current state from a resource generation system and associated resource markets. The model can then compute a policy based on the state as well physical and technical constraints. The policy defines various actions that direct operation of the resource generation system such as resource production and dispatch to markets. Applying the policy to the resource generation results in a modified state which the model extracts along with a measure of optimality which quantifies the success of the policy. Based on these metrics, the model can generate an updated iteration of the policy defining a different set of actions. In this way, the model can gradually develop an optimal policy for controlling the resource generation system.

Multi-market resource dispatch optimization

A traceability system for a bulk commodity supply chain is provided. The system includes a tracking device, a location determination subsystem, and at least one computing device having at least one processor. The location determination subsystem is configured to determine positional information of the tracking device while placed in a bulk commodity traveling along the bulk commodity supply chain. The processor receives the positional information from the location subsystem, extracts positional values from the positional information, and processes the positional values to identify motion primitives. A modeling tool is applied to the identified motion primitives to produce a positional path of the tracking device, which is output, for example, via a user interface. The positional path represents travel of the bulk commodity along the supply chain.

Traceability system for bulk commodity supply chain

A computing system configured to execute a predictive program is provided. The predictive program, in a run-time phase, receives a current value for a remotely sourced forecast as run-time input into an artificial intelligence model. The artificial intelligence model has been trained on training data including a time series of locally sourced measurements for a parameter and a time series of remotely sourced forecast data for the parameter. The predictive program outputs a predicted forecast offset between the current value of a remotely sourced forecast and a future locally sourced measurement for the parameter. The predictive program outputs from the artificial intelligence model a predicted forecast offset based on the run-time input.

Predicted forecast offset from remote location sensor

Growing community of inventors

Seattle, WA, United States of America

Peeyush Kumar