Optimization of the placement of banknotes in ATMs

    Inside, an average ATM has 1 to 3 million rubles . They are arranged in 4 cassettes, each of which contains banknotes of its own denomination. Replacing the cartridges is similar to replacing the cartridges in the printer: the module is removed (and it does not matter how much money is left - collectors do not even know this by safety rules), and another module is inserted in its place exactly in the direction of the arrow indicating which side and how to do it.

    Collection is an expensive and quite risky procedure in terms of security, so any ATM service is quite expensive. The natural desire of the bank is to reduce the number of collections. It turns out the classic problem of resource allocation: on the one hand, the ATM should always have money for customers, on the other - ideally, the bank wants the last bill to leave the box exactly before collection.

    Banks optimize all processes of working with cash: this is the mentioned modularity of cartridges (so as not to take out money and not be counted by hands), and a special arrangement of bags in the car, exactly corresponding to their order of use along the route, and much more. At the same time, everyone perfectly understands that if the ATM does not give money, the client will be upset: this is not a beer stall where you can choose another variety or just cross the road and get the goods from another place. Not receiving money from a "native" ATM, the client is very upset, indignant and begins to lose confidence in the bank.

    Accordingly, the main priority is to always have money. At the same time, I want them to be, if not end to end, then with a minimum margin - and it is here that we need very good IT technologies that can cost very good money.

    Tasks


    In principle, analytics has two tasks :
    1. It is optimal to formulate a banknote breakdown: which bills, by what value and by how many pieces should be in the ATM.
    2. Know the collection schedule and calculate, depending on the specific situation, such a supply of funds so that it lasts until the next time without unscheduled departures.

    Let's start with a breakdown of bills . As I already said, there are usually 4 cassettes there: it’s understandable, the thing is that two of them immediately go under the small ones: under hundred and five hundred. But there are nuances: for example, 500-ruble bills can not be put at all, but put more than others. Or use two boxes at once under bills of the same denomination. Or don’t put five thousandths and so on. Here it all depends on the consumption scenario and the issuance algorithm: if this is a student canteen, then collection may be needed with 3 full cassettes with run-out hundredths.

    Second moment- this is actually how much balance should be. The scenario for using an ATM is important here: for example, there are exclusively salary ATMs that are in offices, there are ATMs that are in shopping centers and there are ATMs that are located in squares at train stations. Obviously, salaries will have pronounced “critical days”, those in shopping centers will have seasonal peaks and declines corresponding to general peaks and declines in retail, in squares and train stations, much depends on fluctuations in the density of the stream and its nature.

    For example, there is an ATM in a business center. In those days when there is a salary in offices, it must be clogged to the brim with large bills, so that those who remove can withdraw as much as necessary. Then, for example, it is filled by a third with bills of lesser denomination, and by some holiday or weekend - it is filled more. Based on the analysis of transactions, you can understand the peaks of withdrawals, understand the customer profile of this ATM, find out what kind of breakdown you need to do.

    There are a lot of nuances, but in general, the task is largely reduced to solving the problem of providing goods on the shelf for shops . For all this, there are a bunch of security features, for example, a ban on the systematic visit of an ATM (using the same routes).

    For optimization we need to know


    About ATM:
    • Number of cassettes
    • Delivery algorithm (can I choose large / small?)
    • Does he accept money or only gives
    • What additional services does he provide
    • How and when did you withdraw money (withdrawal history and balances for Data Mining)
    About what's around the ATM:
    • Where does he stand (in a business center, in a shopping center, in a bank branch)
    • The operating mode of the place where it stands
    About the conditions of the task:
    • How can ATMs be combined into groups?
    • Restrictions on the breakdown of bills (50 bills ready to put? And 100?)
    • Multi-currency ATM or not?
    • The cost of finding money at an ATM
    • The cost of collection
    • Purpose (i.e. what level of service is needed)


    Decision


    The first part of the solution is architectural . You do not need to figure out how to carry the money, but first you need to design and build an ATM network so that it works in an optimal way. This is done based on a heap of data. Our analytics can plan the ATM network itself in such a way that everything optimally takes into account many restrictions - the presence of a number of competitors, the patency of streets, shopping centers, train stations, seasonal changes, and so on.

    Then we must decide how much money and in what breakdown should be lying in the ATM. Plus - how often it should be collected.

    ATMs can also be replenished by customers, and money is collected there in a separate cassette, which can also be overloaded. No money is issued from this cassette: input-output at an ATM in different processes. From the point of view of IT, it seems logical to use this money for extradition, but according to the law, all this must first be done through bookkeeping, and for this, the cassette must be brought to the bank branch. Therefore, it is advantageous to immediately make a replacement of a cassette with money from replenishment for collection (since we have arrived anyway).

    Next, you need to ensure the correct filling of ATMs. To do this, we take a system that allows us to optimize the route from the point of view of minimizing costs and at the same time set restrictions like “routes should change once a week” (except for security services, no one knows which route the collector's car will go tomorrow).

    Returning to logistics, we have several blocks: the first - we need to strategically plan it, the second - we need to understand who to collect with what amounts, and so on, and the third - we need to optimize the collection schedule, both for individual ATMs and for ATM chains at once. If there are a lot of them, then for each of us we consider how to optimally collect it, and then we group them by route.

    For example, if one ATM needs to be collected once every 2 days, and the other once every 3 days, this will be two departures. If you collect both at once, then it will be one trip, which is more profitable and safer. At this level, the solution is to put in less cash to those ATMs that need to be collected less frequently, so as not to create unnecessary idle stocks.

    It is clear that all of these blocks can be done on anything. For example, logistics is Oracle Transportation Management, forecasts for IBM SPSS to do, the deployment of the network for something else. Almost every major vendor has its own solutions.

    Reality


    The main obstacle to good optimization is security requirements. To put it simply, optimization itself implies a clear system of visiting points, and protection requirements say that such a system does not last long. Further there are options:
    • The security service itself works out the routes . We can just make recommendations and make monitors how much money is in which ATM: connecting ATM software with the central office is a common thing. But we are just solving another problem - we do not want to react reactively: “Ahh, our red light is on, the remainder is zero, panic!”, We want to either help this red light come on today and notify that there is only enough money for tomorrow - or we must make a competent collection plan so that the illumination of this red light bulb is something extraordinary for us. Roughly speaking, it is much more economical to spend energy on solving the problem at the architectural level than to try to correct the situation at the operational level each time.
    • The security service is ready for innovation . In this case, an environment usually develops where it is possible to build routes taking into account the optimal traffic pattern.
    • The bank is ready for strategic planning . Here we are talking about miscalculation of the optimal network location, setting optimal routes and alerts about the stock of money.


    Little future


    Given the level of work of predictive software on the example of this reading of the thoughts of call-center clients , one can do the following: when authorizing a card, individual work scenarios and offers can be generated. For example - “top up my phone”, “give me money for lunch”, “to me as usual”, all sorts of “I feel lucky”, combined with the purpose of payments - reminders about paying for the Internet (with an instant opportunity to resolve the issue), paying fines (even dialed 20 minutes ago) and so on.

    The next obvious step is to understand which customers have what to offer. You can react to actions with an ATM, analyze patterns of behavior. For example, launching personalized marketing companies, making “tasty” offers to customers who may leave, and so on.

    In fact, there are no such complex projects in Russia. At best, individual blocks have been worked out. Banks practically do not do this: not everyone has the idea that one can live even better and due to what.

    Saving


    From the standard scheme “when there are 200 thousand left in the ATM, you need to panic and go” to the smart ATM without taking into account space decisions, the difference is from 40 to 80 thousand savings per year on the ATM. Given the number of ATMs of one bank, this difference fully covers the cost of solving the problem and introducing the system.

    Also popular now: