Techniques for programmatically updating the status of assets tracked in a supply chain environment

1. A method, comprising:

obtaining, by one or more processors of a device, sensor data collected from a tracking device associated with a first asset, wherein the sensor data includes indirect sensor data, and wherein the indirect sensor data includes at least one of environmental and temperature data;

determining, by the one or more processors of the device, asset workflow information associated with the first asset, wherein the asset workflow information indicates an expected workflow associated with the first asset;

training an intelligence engine configured to generate an inference of a status associated with the first asset;

using, by the one or more processors of the device, the intelligence engine to generate an inferred state based on the indirect sensor data and the asset workflow information;

determining, by the one or more processors of the device, the status associated with the first asset based on the inferred state;

storing, by the one or more processors of the device, the status of the first asset in an asset tracking database;

determining historical sensor data associated with the first asset, wherein the historical sensor data includes direct sensor data;

determining, by the one or more processors of the device and based on the historical sensor data, ground truth data for the intelligence engine;

comparing the inferred state and the ground truth data; and

re-training, by the one or more processors of the device, the intelligence engine based on comparing the inferred state and the ground truth data.

2. The method of claim 1 , further comprising:

providing, by the one or more processors of the device, the status of the first asset to one or more downstream systems as a notification or event.

3. The method of claim 1 , wherein:

the ground truth data comprises customer feedback.

4. The method of claim 1 , further comprising automatically unpairing the tracking device form the asset.

5. The method of claim 1 , wherein the first sensor data comprises environmental sensor data that lacks GPS sensor data and is used to determine a geolocation-related status of the asset.

6. The method of claim 5 , wherein determining the asset workflow information is based on metadata associated with the first asset.

7. A system, comprising:

at least one processor; and

at least one memory storing computer-executable instructions, that when executed by the at least one processor, cause the at least one processor to:

obtain sensor data collected from a first tracking device associated with a first asset, wherein the sensor data includes indirect sensor data, and wherein the indirect sensor data includes at least one of environmental and temperature data;

determine asset workflow information associated with the first asset, wherein the asset workflow information indicates an expected workflow associated with the first asset;

train an intelligence engine configured to generate an inference of a status associated with the first asset;

use the intelligence engine to generate an inferred state based on the indirect sensor data and the asset workflow information;

determine the status associated with the first asset based on the inferred state;

store the status of the first asset in an asset tracking database;

determine historical sensor data associated with the first asset, wherein the historical sensor data includes direct sensor data;

determine, based on the historical sensor data, ground truth data for the intelligence engine;

compare the inferred state and the ground truth data; and

re-train the intelligence engine based on comparing the inferred state and the ground truth data.

8. The system of claim 7 , wherein the at least one processor are further configured to:

provide the status of the first asset to one or more downstream systems as a notification or event.

9. The system of claim 7 , wherein:

the asset workflow information is encoded as a finite state machine (FSM).

10. The system of claim 7 , wherein:

a first portion of the historical sensor data corresponding to second indirect sensor data is used as training input data; and

a second portion of the historical sensor data corresponding to direct sensor data is used as the ground truth data.

11. The system of claim 10 , wherein:

the inferred state is associated with geolocation of the first asset; and

the direct sensor data includes at least one of GPS or other locationing receiver data and manual verification data.

12. The system of claim 11 , wherein:

the indirect sensor data includes environmental sensor data.

13. A non-transitory computer-readable medium including computer-executable instructions stored thereon, which when executed by at least one processors, cause the at least one processors to:

obtain first sensor data collected from a first tracking device associated with a first asset, wherein the first sensor data includes indirect sensor data, and wherein the indirect sensor data includes at least one of environmental and temperature data;

determine asset workflow information associated with the first asset, wherein the asset workflow information indicates an expected workflow associated with the first asset;

train an intelligence engine configured to generate an inference of a status associated with the first asset;

use the intelligence engine to generate an inferred state based on the indirect sensor data and the asset workflow information;

determine an intelligence engine configured to generate an inference of a status associated with the first asset;

determine the status associated with the first asset based on the inferred state; and

store the status of the first asset in an asset tracking database;

determine historical sensor data associated with the first asset, wherein the historical sensor data includes direct sensor data;

determine, based on the historical sensor data, ground truth data for the intelligence engine;

compare the inferred state and the ground truth data; and

re-train the intelligence engine based on comparing the inferred state and the ground truth data.

14. The non-transitory computer-readable medium of claim 13 , wherein the at least one processors are further configured to:

provide the status of the first asset to one or more downstream systems as a notification or event.

15. The non-transitory computer-readable medium of claim 13 , wherein:

the asset workflow information is encoded as a finite state machine (FSM).

16. The non-transitory computer-readable medium of claim 13 , wherein:

a first portion of the historical sensor data corresponding to second indirect sensor data is used as training input data; and

a second portion of the historical sensor data corresponding to direct sensor data is used as ground truth data.