Location-based presence model for item delivery

Abstract

Methods and systems for prescription drug shipping selection are provided. The methods and systems include operations comprising: obtaining, by a server, activity data from a plurality of devices associated with a location, the activity data representing different types of activities that take place at the location over a threshold period of time; aggregating, by the server, the activity data to generate a location-based presence model for the location, the location-based presence model indicating likelihoods that a person is present at the location at a plurality of different time windows; and identifying, by the server, based on the location-based presence model, a time window for delivery of a perishable item to the location.

What is claimed is: 
     
         1 . A system comprising:
 a memory element;   a wireless network interface, configured to transmit and receive communications over a network; and   at least one processor communicatively coupled to the memory element and the wireless network interface, the at least one processor configured to:
 transmit a graphical user interface (GUI) for presentation by a user device, via the wireless network interface, wherein the GUI includes a listing of a plurality of devices associated with a location; 
 receive user input selecting a subset of the plurality of devices permitted to share activity data, via the GUI; 
 establish wireless communications connections over a wireless network to the subset of the plurality of devices, via the wireless network interface; 
 continuously obtain the activity data from the subset of the plurality of devices during a modeling period of time, via the wireless network interface, wherein the activity data indicates human interaction or lack of human interaction with the subset of the plurality of devices at the location during the modeling period of time; 
 aggregate the activity data for the modeling period of time, to generate a set of aggregated activity data; 
 model a location-based presence model for the location based on the set of aggregated activity data, the location-based presence model indicating a likelihood that a person is present at the location during a plurality of time windows; 
 identify a time window from the plurality of time windows for delivery of a perishable item to the location, based on the location-based presence model; and 
 transmit an updated GUI for presentation by the user device, via the wireless network interface, wherein the updated GUI presents the time window for delivery of the perishable item. 
   
     
     
         2 . The system of  claim 1 , wherein the at least one processor is further configured to:
 transmit a second updated GUI for presentation by the user device, via the wireless network interface, wherein the second updated GUI presents the plurality of time windows; and   receive second user input selecting a user-selected time window of the plurality of time windows for delivery of the perishable item, wherein the time window for delivery comprises the user-selected one of the plurality of time windows.   
     
     
         3 . The system of  claim 1 , wherein the at least one processor is further configured to:
 identify one of the plurality of time windows, based on a highest likelihood that a person is present at the location during the plurality of time windows, wherein the one of the plurality of time windows comprises the time window for delivery.   
     
     
         4 . The system of  claim 1 , wherein the at least one processor is further configured to:
 obtain machine learning input data, via the wireless network interface, wherein the machine learning input data includes at least one of: sequential activity for the plurality of devices in a particular sequence, concurrent activity or consecutive activity for a first one of the plurality of devices within a threshold time period of a second one of the plurality of devices, no detected activity for the plurality of devices, majority activity for a majority of the plurality of devices, minority activity for a minority of the plurality of devices, and all activity for an entirety of the plurality of devices; and   use machine learning to generate the location-based presence model based on the machine learning input data, wherein the aggregated activity data comprises the machine learning input data.   
     
     
         5 . The system of  claim 1 , wherein the at least one processor is further configured to:
 rank the plurality of time windows based on the location-based presence model, to generate a ranked list of potential time windows for delivery of the perishable item;   transmit the ranked list of potential time windows for presentation by the user device, via the wireless network interface; and   identify the time window for delivery by receiving second user input selecting the time window from the ranked list of potential time windows.   
     
     
         6 . The system of  claim 5 , wherein the at least one processor is further configured to:
 transmit a preferred one of the ranked list as a highlighted entry in the ranked list, for presentation by the user device, wherein the preferred one is associated with a highest likelihood that a person is present at the location during the plurality of time windows.   
     
     
         7 . The system of  claim 1 , wherein the at least one processor is further configured to:
 receive second user input selecting a low level of data sharing for the subset of the plurality of devices, via the GUI; and   obtain low level activity data based on the second user input, via the wireless network interface, wherein the low level activity data includes times of operations and types of operations associated with the subset of the plurality of devices, and wherein the activity data comprises the low level activity data.   
     
     
         8 . The system of  claim 1 , wherein the at least one processor is further configured to:
 receive second user input selecting a high level of data sharing for the subset of the plurality of devices, via the GUI; and   obtain high level activity data based on the second user input, via the wireless network interface, wherein the high level activity data includes low level activity data and at least one of: an identity of a device operator and an image of the device operator, and wherein the activity data comprises the low level activity data and the high level activity data.   
     
     
         9 . The system of  claim 1 , wherein the at least one processor is further configured to:
 obtain a plurality of likelihoods that a person is present at the location during the subset of the plurality of time windows, wherein ones of the plurality of likelihoods is associated with a respective one of the subset; and   when a first likelihood of the plurality of likelihoods exceeds a predetermined threshold,
 determine that the first likelihood indicates a high probability that a person is present at the location during an associated time window, wherein the subset includes the associated time window; and 
 transmit a second updated GUI for presentation by the user device, via the wireless network interface, wherein the second updated GUI presents a list of potential time windows for delivery, including positioning the associated time window in the list to indicate the high probability. 
   
     
     
         10 . The system of  claim 1 , wherein the at least one processor is further configured to:
 obtain a plurality of likelihoods that a person is present at the location during the subset of the plurality of time windows, wherein ones of the plurality of likelihoods is associated with a respective one of the subset; and   when a first likelihood of the plurality of likelihoods does not exceed a predetermined threshold,
 determine that the first likelihood indicates a low probability that a person is present at the location during an associated time window, wherein the subset includes the associated time window; and 
 transmit a second updated GUI for presentation by the user device, via the wireless network interface, wherein the second updated GUI presents a list of potential time windows for delivery, including positioning the associated time window in the list to indicate the low probability. 
   
     
     
         11 . A method comprising:
 presenting a graphical user interface (GUI) including a listing of a plurality of devices associated with a location, by at least one processor via a communicatively coupled display device;   receiving user input selecting a subset of the plurality of devices permitted to share activity data, by the at least one processor via the GUI;   establishing wireless communications connections over a wireless network to the subset of the plurality of devices, by the at least one processor via a wireless network interface;   continuously obtaining the activity data from the subset of the plurality of devices during a modeling period of time, by the at least one processor via the wireless network interface, wherein the activity data indicates human interaction or lack of human interaction with the subset of the plurality of devices at the location during the modeling period of time;   aggregating the activity data for the modeling period of time, by the at least one processor, to generate a set of aggregated activity data;   generating a location-based presence model for the location based on the set of aggregated activity data, by the at least one processor, the location-based presence model indicating a likelihood that a person is present at the location during a plurality of time windows;   identifying a time window from the plurality of time windows for delivery of a perishable item to the location, by the at least one processor, based on the location-based presence model; and   updating the GUI to present the time window for delivery of the perishable item, by the at least one processor via the communicatively coupled display device.   
     
     
         12 . The method of  claim 11 , further comprising:
 updating the GUI to present the plurality of time windows;   receiving second user input selecting a user-selected one of the plurality of time windows for delivery of the perishable item; and   updating the GUI to present the user-selected one of the plurality of time windows, by the at least one processor via the communicatively coupled display device, wherein the time window for delivery comprises the user-selected one.   
     
     
         13 . The method of  claim 11 , further comprising:
 identifying one of the plurality of time windows, by the at least one processor, based on a highest likelihood that a person is present at the location during the plurality of time windows, wherein the one of the plurality of time windows comprises the time window for delivery.   
     
     
         14 . The method of  claim 11 , further comprising:
 obtaining machine learning input data, including at least one of: sequential activity for the plurality of devices in a particular sequence, concurrent activity or consecutive activity for a first one of the plurality of devices within a threshold time period of a second one of the plurality of devices, no detected activity for the plurality of devices, majority activity for a majority of the plurality of devices, minority activity for a minority of the plurality of devices, and all activity for an entirety of the plurality of devices; and   using machine learning to generate the location-based presence model based on the machine learning input data, wherein the aggregated activity data comprises the machine learning input data.   
     
     
         15 . The method of  claim 11 , further comprising:
 ranking the plurality of time windows based on the location-based presence model to generate a ranked list of potential time windows for delivery of the perishable item, by the at least one processor;   presenting the ranked list of potential time windows, by the at least one processor via the GUI; and   identifying the time window for delivery by receiving second user input selecting the time window from the ranked list of potential time windows.   
     
     
         16 . The method of  claim 15 , further comprising:
 presenting a preferred one of the ranked list of potential time windows as a highlighted entry, wherein the preferred one is associated with a highest likelihood that a person is present at the location during the plurality of time windows.   
     
     
         17 . The method of  claim 11 , further comprising:
 receiving second user input selecting a low level of data sharing for the subset of the plurality of devices, by the at least one processor via the GUI; and   obtaining low level activity data based on the second user input, by the at least one processor, wherein the low level activity data includes times of operations and types of operations associated with the subset of the plurality of devices, and wherein the activity data comprises the low level activity data.   
     
     
         18 . The method of  claim 11 , further comprising:
 receiving second user input selecting a high level of data sharing for the subset of the plurality of devices, by the at least one processor via the GUI; and   obtaining high level activity data based on the second user input, by the at least one processor, wherein the high level activity data includes low level activity data and at least one of: an identity of a device operator and an image of the device operator, and wherein the activity data comprises the low level activity data and the high level activity data.   
     
     
         19 . The method of  claim 11 , further comprising:
 obtaining a plurality of likelihoods that a person is present at the location during the subset of the plurality of time windows, by the at least one processor, wherein each of the plurality of likelihoods is associated with a respective one of the subset; and   when a first likelihood of the plurality of likelihoods exceeds a predetermined threshold,
 determining that the first likelihood indicates a high probability that a person is present at the location during an associated time window, wherein the subset includes the associated time window; and 
 updating the GUI to present a list of potential time windows for delivery, including positioning the associated time window in the list to indicate the high probability. 
   
     
     
         20 . The method of  claim 11 , further comprising:
 obtaining a plurality of likelihoods that a person is present at the location during the subset of the plurality of time windows, by the at least one processor, wherein each of the plurality of likelihoods is associated with a respective one of the subset;   determining whether a first likelihood of the plurality of likelihoods indicates a low probability, a medium probability, or a high probability that a person is present at the location during an associated time window, wherein the subset includes the associated time window; and   
       updating the GUI to present a list of potential time windows for delivery, including positioning the associated time window in the list to indicate the low probability, the medium probability, or the high probability.