Amazon Technologies, Inc.

Sas Institute Inc.

Ivan Borges Oliveira

Manik Kumar

Philip Mark Kaminsky

Phillip Oliver Kriett

Akhand Pratap Singh

Georgios Patsakis

Timothy Lee Jacobs

Susan Edwards Haller

Robert William Pratt

Xunlei Wu

Karthik Ramakrishnan

Manoj Keshavmurthi Chari

Hamza Mustafa Ghadyali

Fang Liang

Natalia Summerville

David Rawlins Duling

Ian Ward

Burak Karacik

Dmitriy Belyi

Daniel Thomas Meiller

Ravikanth Kundurthi

Golbarg Tutunchi

Andrew V Goldberg

Sharmin Pathan

Quico Pepijn Spaen

Diwakar Tiwari

Vighnesh Naik

Golbarg Kazemi Tutunchi

Quico Pepjin Spaen

Scott Shuler

Sendhil Nathan

Suresh Rangaswamy

Pradeep Chowdary Mandava

Techniques for allocating resources and generating resource allocation instructions are described herein. A model can generate a historical error for resources of a facility based on a first set of predicted resource imbalances and historical data for resources imbalances of the facility for a first set of previous time periods. The model can receive a second predicted resource imbalance for the resources of the facility associated with a first future time period. The model can receive real-time signals for states of the resources of the facility associated with a present time period. The model can generate an adjusted resource prediction for the resources of the facility based at least in part on the historical error, the second predicted resource imbalance, and the real-time resource signals. The adjusted resource prediction can be associated with the first future time period.

Techniques for updating resource predictions and resource allocations

Transportation handoffs for increased network density are described herein. In an example, a computer system generates first data indicating a first candidate network path associated with transporting a first load of items between an origin and a destination within a network. The first candidate network path directly connects the origin and the destination. The computer system generates second data indicating a second candidate network path indirectly connecting the origin and the destination. The computer system generates a first input including the first data and the second data to an optimization model and a second input about a second load of items to the optimization model. The computer system determines an output of the optimization model indicating, for the first load, a selected network path a resource assignment per segment of the selected network path. The computer system causes, based on the output of the optimization model, utilization of transportation resources.

Transportation handoffs for increased network density

One challenge for middle-mile route planning is that the set of loads changes significantly between daily planning and execution. Systems and methods are provided for optimizing a transportation plan for a transportation network based on these load changes. The disclosed system re-optimizes a solution by starting from a previously existing plan and previously generated columns (e.g., candidate routes). The disclosed techniques significantly improve the compute time of the system to generate transportation plans that are optimized accordingly to an optimization parameter. The system takes into account the current execution status associated with a given entry of the plan to determine whether the entry should be re-optimized. Entries corresponding to tours that have already commenced, may be at least partially ignored for re-optimization consideration. The disclosed techniques enable state-aware, adaptive re-optimization for even tours that are in-progress or have been tendered.

Techniques for reactive route planning

Techniques for planning resources using block and route information are described. In an example, a computing system determines a demand for item transportation expected during a planning horizon. The computing system determines information about a pre-planned transportation resource available during the planning horizon and costs associated with the pre-planned transportation resource. The computing system uses an optimization model to determine a block having a time length, a tour to transport, during the block, a first portion of the demand using the pre-planned transportation resource, and a second portion of the demand to be transported using an on-demand transportation resource. The computing system indicates, to a first computing device of the pre-planned transportation resource, an assignment of the block to the pre-planned transportation resource.

Resource planning using block and route information

A computing system obtains image data capturing first and second objects. The system determines, based on user-identified data points, boundaries of the objects and generates a component of a dataset by computing a first data value related to an attribute of a key point in the first image; and computing a second data value related to an attribute of a key point in the first image. The system generates a second component of the dataset, the second component representing updated relative information between the first and second object by generating predicted changes in the first data value and second data value for the second image. The system computes a third data value and a fourth data value related to respective data points in a first and second polygon in the second image. The generating the updated relative information is based on the predicted changes and computed values.

Object and data point tracking to control system in operation

Managing the amount of computing resources required to execute a process for determining values of a parameter associated with an object over a lifetime of the object is disclosed here. In one example, a data structure is generated. The data structure including candidate values for the parameter that comply with constraints assigned to multiple dates occurring during the lifetime of the object. The data structure is pruned by aggregating actionable periods. A first combination of candidate values associated with the aggregated actionable periods is determined that results in the minimum amount of the object being provided to the users during the lifetime. A second combination of candidate values associated with the aggregated actionable periods is determined that satisfies a return objective. The second combination of values are usable by a remote computing device to implement a value schedule for the object.

Managing object values and resource consumption

Computer-implemented systems and methods are provided for generating bins for a scorecard. An approximate set of bins is generated by applying an optimization model to binning data. The optimization model includes an objective function, constraints, and surrogate weight of evidence metric(s). The approximated set of bins are then used in scorecard operations.

Constrained optimized binning for scorecards

Growing community of inventors

Bellevue, WA, United States of America

Ivan Borges Oliveira