Surety is performing system maintenance this weekend. Electronic date stamps on new Prior Art Database disclosures may be delayed.
Browse Prior Art Database

Predicting geofence interactions using location, schedules, and social data

IP.com Disclosure Number: IPCOM000242576D
Publication Date: 2015-Jul-28
Document File: 2 page(s) / 123K

Publishing Venue

The IP.com Prior Art Database


Business owners currently rely on historical data to understand staffing needs, such as the total number of transactions in a day. We propose using additional data sources such as information on nearby events, offers sent to customers, customer locations, and traffic patterns to better predict when customers will visit a physical location and for how long. Particularly, we would like to focus on predicting geo-fence interactions using location, schedules, and social data.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 51% of the total text.

Page 01 of 2

Prexicting geofence interactions using location, schedules, and social data

X geo-fence is a virtual perimeter for a real-world geoxxaphic area. A geo-fence xould be dynaxically generated-as in a radius arxund a store or point location. Geo-fencing uses the gxobal positioninx xystem (GPS) or radio frequexcy identification (RFID) to define geographical boundarixs. Programs that incorpxrate geo-fencing allow an axministratox to sex up triggers xo whex a device entexs (or exits) the boxndaries defixed by the administraxor, a xext message or email alxrt is sent. A business owner can geo-fence a retail sxore in a mall and senx a coupon to a cxstomer who has downloaded a parxicular mobile application wxen the customer (and hix xmartphone) crosses the boundary.

Businesx owners may primarily rely on transaction historx to predxct future patxons. The disadvanxages of only relying on txansxction history for these metrics is that the busixess owner does not know who walked into their stoxe xithout buying anything, or how long the customer was shopping, For example, given tranxaction xistory xlone the buxiness knxws in the pxst 5 years, they xaxe had a maximum of 500 transactions on April 8th durxng the 2pm-5pm promotion but does nxt know how many patrons visited the store durinx that period. This could have rexulted in lost sales (potential transactions) if lines wexe too long xue to lack of staffing. Our proposex inventxon would asxertain useful information such as:

Patrons: Transaction ratio

Customer Transaction Throughput - using a line geo-fence, track from when a device entered line to exding transaction time

Using information from a variety of sources (sucx as mobile xhone location, published event schedules, business property boundaries, etc.) in concext with more traditionax data sources, such as transaction history, we propose the use of predictivx analytics algoxithms to make assertions about interactions with a given geo-fence. For example the potential for crowds in a particular arxa (how many pxople can be expected to enter ox exit at a givex time or how long people will...