Browse Prior Art Database

A Process for Optimal Cash Deployment to a Network of Automatic Teller Machines

IP.com Disclosure Number: IPCOM000022094D
Original Publication Date: 2004-Feb-24
Included in the Prior Art Database: 2004-Feb-24
Document File: 3 page(s) / 120K

Publishing Venue

IBM

Abstract

A system / business process designed to schedule deliveries of cash to an automatic teller network

This text was extracted from a PDF file.
At least one non-text object (such as an image or picture) has been suppressed.
This is the abbreviated version, containing approximately 54% of the total text.

Page 1 of 3

A Process for Optimal Cash Deployment to a Network of Automatic Teller Machines Overview

Managing a network of automatic teller machines has typically been a labour intensive business. This patent describes a method for programmatically managing a large network of machines with minimal human intervention.

Automatic Teller Machine

Consolidated View

Reconciliation Rejects etc

Automatic Teller Machine

Automatic Teller Machine

Automatic Teller Machine

Automatic Teller Machine

 Select Machines For

 Host system

Predict Cash requirements

Manual Override

Fax or email to security co

The process consists of:
1. Collecting data on each machines daily-dispensed record.
2. Selecting machines on a daily basis for replenishment via a calendar system.
3. An optimal method for forecasting the next replenishment amount
4. Generating suitable documentation to instruct the security company
5. Reconciling rejected cash, notes of poor quality etc.

Additionally the process has to be capable of handling special or exceptional events, e.g. early replenishments due to a run on cash at a certain machine or handling of public holidays.

1

[This page contains 4 pictures or other non-text objects]

Page 2 of 3

1. Collecting data on each machines daily-dispensed record

The automatic teller machine generates a transaction record to its host system once per day reporting the total cash remaining and the total cash dispensed. This information permits the current cash levels to be tracked on an on-going basis. These records provide the basis for generating a forecast for machines selected for replenishment on a particular day. This information has to be collated into a central database, which contains a record per machine.

2. Selecting machines on a daily basis

Machines are selected using a calendar mechanism this process takes into account various key lead times:
* Replenishment cycle of ATM (every n days, twice weekly, etc)
* Lead time required by the security firm to build cartridges of cash for machines
* Lead time required by security firm to deliver cash to machines

This provides a unique replenishment profile for a group of machines.

The machines, which have reached the calendar event for replenishment, are selected.

           Build Cartridge Deliver Cartridge Machines needing
Completion date to start of cycle

3. An optimal method of forecasting

A method known as time series analysis is used to predict the next replenishment for the selected machines. The algorithm used is published technique called "simple exponential smoothing". Given by the equation:

1

       
t
ααLYt L

1

=+-( - t

)

F k t =

L

+

The term L of t is the new level, at time t. Alpha is the respective weights, it is analogous to the span in moving averages. With simplification, substitution and testing we have derived the following equation from the root equation described above.

1

0.70.1 3 -

   =- + t Ld...