Target Brands, Inc.

Tikhon Jelvis

Sayon Majumdar

Zeynep Erkin Baz

Michael Rorro

Saibal Bhattacharya

Bryce Cooks

Ashwin Rao

Michael Staab

Bo Yang

Chandrashekar Tallichetty

Yu Gao

Adam Riggall

Duane Sizemore

Jeffrey Eul

Peter Kim

Abhijith Neerkaje

Moritz Drexl

Jun Qin

Aron Gruzman

Omar N Jabri

Cory Morris

Gregory Malecha

Kaveh Khodjasteh Lakelayeh

Shaokuan Chen

Yang Liu

Kaveh Khodjasteh

Joseph Cox

Methods and systems are described for providing hierarchical demand forecasting for state space reduction. Using a hierarchical architecture, a base model may be trained to capture a range of shared structure in a first data set that can be used to draw inferences on using smaller sets of data representative of the 'whole picture.' For example, training data may be sampled and prepared and used to train a base model in a first stage. In a next stage, one or more downstream models may be trained on the structure and samples of uncensored demand generated by the base model to produce forecasts for items and locations, including items and locations for which there may be little or no historical data. The downstream models that would otherwise require a large amount of data for training can be generated on demand using less training data, training time, computing processing, and memory.

Hierarchical demand forecasting modeling

Methods and systems for forecasting demand for items across multiple channels are disclosed. In some implementations, multi-channel demand forecasting may be performed on a per-item, per-location basis, by selectively generating item-location forecasts for each item and location within a supply chain for each channel, or disaggregating a chain level forecast on a per-item basis to each location. Particular selection of an appropriate model, and selective training of models, allows for efficient computation of such forecasts across a large supply chain with thousands of locations and hundreds of thousands, or millions, of items for which forecasts are generated.

Multi-channel demand planning for inventory planning and control

An interactive visualization tool for displaying metrics associated with inventory related output from a replenishment simulation for an enterprise supply chain is described. Predicted values for metrics associated with simulation output may be received on a per item, per node, per epoch basis for a plurality of items associated with the supply chain at a plurality of nodes of the supply chain over a plurality of epochs. A visualization tool user interface may be rendered that includes item and node menus enabling user selection of varying granularity levels for viewing the metrics. The metrics may be aggregated according to a presently selected level of granularity, and represented within a single graphical view of the user interface to visualize projected inventory positions of items across nodes of the supply chain. The aggregated metrics may include an expected projection and a worst-case projection for one or more types of the metrics.

User interface for visualizing output from supply chain replenishment simulation

Methods and systems for managing inventory items within a supply chain are disclosed. One method includes receiving, at a software tool, inputs related to a plurality of inventory items, the inputs including a cost of holding each of the plurality of inventory items at a location type. The method includes determining, individually for each inventory item of a plurality of inventory items, an optimal inventory balance across a plurality of locations, wherein the optimal inventory balance is a predetermined statistical availability level set based on a desired customer availability of the inventory item. The method further includes automatically generating one or more inventory adjustment requests to achieve the optimal inventory balance across each of the plurality of locations for each of the plurality of inventory items.

Inventory management

Event-based replenishment simulation for an enterprise supply chain is described. On a per item, per node, per epoch basis, the simulation may generate a stream of action events based on forecasted demand, supply chain logic, and policy inputs that are applied to a current-run state of the supply chain in order to yield a stream of observation events. Requested metrics may be received, and the observation events may then be transformed to predict values for the metrics as output of the simulation. The simulation may be repeated for a given epoch using discrete demand values from a demand distribution, for a plurality of epochs, and/or across a plurality of items at a plurality of nodes. Resultantly, the simulation output can be used for predicting a future run-state of the supply chain across items and nodes.

Supply chain replenishment simulation

In some implementations, a method performed by data processing apparatuses includes receiving order data that represents a plurality of ordered items for delivery to a location; selecting a first policy from a store of first policies; transforming at least a portion of the order data into a plurality of item units, based on rules associated with the selected first policy; selecting a second policy from a store of second policies; for each item unit, based on rules associated with the selected second policy, modifying the item unit to include annotated information that corresponds to operations to be performed on the item unit; and generating instructions for grouping the plurality of item units for delivery to the location, based on the annotated information for the item units.

Platform using instruction engine to simulate and perform warehouse processes

Growing community of inventors

Sunnyvale, CA, United States of America

Tikhon Jelvis