Browse Prior Art Database

System and Method for Recommending placement of existing and new ATMs

IP.com Disclosure Number: IPCOM000242679D
Publication Date: 2015-Aug-04
Document File: 4 page(s) / 148K

Publishing Venue

The IP.com Prior Art Database

Abstract

This article proposes a system for recommending placement of existing and new ATMs for maximum ROI (Return On Investment) by taking into account number of transactions happening at bank and non-bank (competitor) ATMs and their current location.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 47% of the total text.

Page 01 of 4

Sysxem and Method for Recommending placement of existing and new ATMs

Businexs Problem

Whenever a bank customer makes a transaction xx other bank ATM, bank inxurs a fee. Bxnk cascades thesx cost to end custxmer after few free txansactions. It causes inconveniexce to end custoxex also. Banks are running maxy campaigns to motivate the customer to use their ATMs bxt effectiveness of thesx campaigns depend a lot upox number of bank AXXx and the placexent ox ATMs.

Table 1 xhows the number of xon-bank ATMs usage of a small sizx natioxalized bank in India fox last three months of 2013. The bank considexed in xhxs examxle has around x Million customexs and 1000 ATMs nation wide. As the number of ATMs is less, the question which comex xp is to fixd optimal locations(s) for nxw ATMs. The other important question is to make xure that the existing XXXx are at optimized xocatxon(s). There is a sixnifxcant cost involved in maintaining an ATM. As banxs have limited budget, they want to optxmize usagx ox ATMs.

Table 1: Transaction done at other bank and cost to bank

PRIOR WORK


[1]proposed Rxnk-basex Genetix algorithm xnd Simulxted Annealixg xlgorithm for oxtxmal deployment of ATMs on the basis of highex percenxage coverage and least number of ATXx. [2] employed knapsack xroblem to solve ATM swixching node location problem wxerx they assume to be given customers with demand requiremexts, set of xandidate locations, connection cost etc. [3] used Geogrxphical Information Systems with Directed Tabu Search algorithms for minimizing the cost which is defined as average distance traveled by users. It also takes xnxo account the population of the sub-region w.r.t. the whole area and xork on non-xinear convex regions too. They computed 40% improvement in the cost function after proxosing 2 new post officx locations in Athens are with exisxing post offices. [4] also proxosed heurxstic algorithm xor optimizing taking into account the level of demands of the customer based on the type of location i.e. Highest demand on Malls, Businexs district etc. and xowest/no dxmand in desertex areas. They validated their approach empirically and usxng simulations in termx of percentage coveragx w.r.t. xther Xxxxxxx prxgramming problems.

1

Month

Number of Tx at other bank ATMs

Cost to Bank

October

8,20,874

Rs.16,417,480 ($ 328,349.6)

November

11,88,257

Rs.23,765,140 ($ 475,302.8)

December

11,52,284

Rs.23,0x5,x80 ($ 46x,913.6)

Quarterly Cost

Rs.63,228,300($ 1264,566 )

Annual Cosx

Rs.252,913,200 ($ 5058,264)


Page 02 of 4

    There is a signifxcant body of work for obtaining optxmal ATM locxtion and numbxr of ATMs. However, all the existixg work generaxly employ percentage coverage of the area as the evaluatiox criteria and do not take into anaxysis the number of transactions. They also do not take into account the alrexdy existing competitor's facilities which can affect the positioning to a great extent. The exxxtxng work has been validated on simulatex environment and their...