GB2410363A - Detecting misplaced cash items within cash drawers - Google Patents

Abstract

In a method of detecting misplacement of cash items in a multi-compartment cash drawer, the drawer has means for measuring the amount of cash in each compartment and the method compares the cash measured as present before and after a transaction to detect any discrepancy resulting from the misplacement. The method includes the calling up of a table of cash movements. The cash drawer has weighing means for the individual compartments to determine the amount of cash in each.

Description

1 2410363

DETECTING MISPLACED CASH ITEMS WITHIN CASH DRAWERS

This invention concerns techniques for detecting misplaced items within cash drawers and for compensating for the effects of such misplacement upon the accuracy of cash s counting. The invention finds particular application in relation to 'intelligent' cash registers having cash drawers that can count the cash they contain.

Cash drawers are usually used in retail transactions, although they may also be used in banks or other cash-handling environments such as payroll or administrative offices.

lo For brevity, the term 'cash drawer' in this specification is to be taken to encompass static cash trays with or without lids: it does not require a movable drawer.

The term 'cash' used herein does not merely refer to banknotes and coins but also to any type of accepted tender or cash item, for example promissory notes, vouchers, tokens and the like. For ease of reference, the foregoing items based upon light flexible substrates like paper or card shall be referred to simply as 'banknotes', and the remainder based upon solid, commonly metallic substrates shall be referred to simply as 'coins' in the remainder of this specification. For present purposes, the main distinction is that individual coins are usually heavier than individual banknotes.

Cash drawers generally use a separate compartment for each denomination of note or coin as well as other payment media such as tokens, coupons or vouchers. However, to reduce the number of compartments required, some applications may employ compartments that each handle more than one denomination.

In retail sectors it is a longstanding problem that the value of cash held within cash registers is not immediately obvious and determining this value accurately requires a time-consuming manual count operation. The value of cash held within cash registers is important to know for management purposes, for example to determine when a cash drawer is full, or nearly full, and an amount of cash (known as a skim) needs to be removed from that drawer. This action is commonly referred to as 'skimming' or lifting' cash from a cash drawer. Cash values are also needed to determine cash levels within a cash register drawer during the day so that cashflow information may be obtained and fraudulent practices may be spotted. In all cases, it is desirable that accurate cash value data is obtained frequently throughout the working day or shift, that this is done with as little intrusion as possible to the operation of the cash register and that as little operator intervention is required as possible. This has led to the development of intelligent cash registers that can count the cash they contain. s

An example of an intelligent cash register is disclosed in the Applicant's co-pending European patent application EP 0724242, which describes a cash register comprising weighing means arranged to take weight readings from the cash compartments of a cash drawer. In conventional fashion, each compartment is assigned to a respective lo denomination of banknote or coin. The weight readings of each compartment may then be converted to a cash value by a processing unit so that the total cash value contained in the register may be obtained without manual intervention.

It is of course inherent in this system that the weight of cash held in each cash compartment is measured and that there is a means for correlating the measured aggregate weight of cash items with an expected weight for each cash item that is expected to be in that compartment. The number of units of that cash item, and hence the value of cash in the compartment, can then in theory be deduced simply by dividing the aggregate weight of items by the expected weight per item. Of course, in practice, deducing the cash value from the weight of a group of cash items is a more complex and challenging task. The ideal of nominally- identical cash items weighing the same as each other and corresponding to a predetermined expected weight is not encountered in real life where cash items become moist, dirty or damaged.

Real life also departs from theory in terms of human behaviour. In an ideal world, cashiers would invariably and unerringly place cash items into the cash compartments in which they are supposed to go. In practice, cashiers work in a hurry in a pressured environment and even if they try to be diligent, which not all cashiers do, they make occasional mistakes. The result is that cash items end up in the wrong compartments and especially that different types of cash item become mixed in a single compartment, particularly at busy times and when the tender includes low value coins or notes.

For example, different types of coin may end up in the same coin compartment.

Different types of banknote may end up in the same banknote compartment. Both scenarios are all the more likely in currency systems where coins and banknotes are of similar colour or size to other coins and banknotes. Coins may even end up in a banknote compartment and vice-versa, most commonly when a coin aimed into a coin compartment misses, bounces or otherwise deflects into an adjacent banknote compartment. A coin misplaced in this way may not be immediately evident because it will tend to be concealed by the banknotes and will migrate toward the bottom of the compartment under gravity, particularly if the cash drawer is of a type that moves l o in and out of the cash register as it is opened and shut. In all cases, the operator's error will cause the cash drawer to miscount and thus generate some form of error signal, although the actual aggregate cash within the cash drawer may be correct.

As an honest error may wrongly be taken to be fraudulent activity requiring disciplinary action or criminal proceedings, it is important to detect and correct the error or better still to avoid it in the first place. Conversely, it is also important that frequent error messages do not obscure genuinely fraudulent activity or mislead managers to ignore fraudulent activity thinking it to be honest but occasionally erroneous operation. In any event, it is not desirable to misplace cash items because the wrong denomination may subsequently be issued as incorrect change to a customer or be passed onwards in error as part of a bank deposit.

Another aspect of the problem is the performance of cashiers. Misplaced items generally indicate that a cashier is not performing properly and it would be useful to monitor the number of error occurrences, potentially as a percentage of the number of transactions undertaken (or even number of items transacted) in order to determine cashier performance and, by training or disciplinary action, to minimise errors at source.

As a simple example, Table 1 sets out the expected and actual number of coins in a cash drawer having six coin compartments, following a coinsonly transaction. A notes-only transaction or a notes-and-coins transaction would follow the same principles. Whilst the example shows only one coin misplaced, it is reasonable to assume that, in real life, more than one coin or note could be misplaced.

In general, if a misplaced coin or note is the cause of a discrepancy, one or more compartments in the cash drawer will contain more units than would be expected from the expected transaction. This excess will be manifested by a higher unit count within the compartment(s) concerned.

I'he example assumes that the cash drawer contains a total of 0.83 units of a given currency made up of five different coins, this accounting for the 'expected' column in which one coin of each respective denomination is expected to be in each of the 0.01, lo 0.02, O.lO, 0.20 and 0.50 compartments. However, the actual measured value is 0.78 because a misplaced 0.10 coin has been registered by the 0.05 compartment rather than by the 0. l 0 compartment.

Table 1

Denomination Expected Actual Number Value Number of of Coins Coins 0. 01 1 1 0.02 1 1 0.05 1 0.10 1 0.20 1 1 0.50 1 1 The result of the misplaced coin will be an error of 0.05 units upon comparison of the expected cash value against the cash value measured by the cash drawer. This is despite the fact that the correct cash value is actually present in the cash drawer as a whole.

In the example of 'able I above, the expected value of the transaction is known. If the expected value matches the measured value of the cash drawer contents, then no further action is required. Any mismatch may be due to a number of circumstances, the main categories being that the cashier has made an erroneous or fraudulent transaction or that the cashier has misplaced one or more coins or notes within the cash drawer.

In general, it is possible to evaluate the expected format of a transaction and compare s it with the actual transaction and identify where an error may have occurred. For example, as a working assumption, cashiers will generally issue change in such a way as to use as few coins (and/or notes) as possible. Using this assumption, it is possible to evaluate the likely denominational composition of any transaction based on the tender proffered by the customer and the transaction value. For example, if the 0 customer proffered 10.50 for a transaction valued at 6.27 and hence received 4.23 in change, the most likely variation within each compartment of the cash drawer would be a shown in Table 2.

Table 2

Denomination Expected Expected Change Format Movement 0.01 1 -1 0.02 1 -1 0.05 0.10 0.20 1 -1 0.50 +1 1.00 4 -4 1 0.00 +1 Other combinations of payment or change could of course arise, but are only likely to arise if the customer specifically requests a particular format of change or if one or more of the denominations used for change is not available. For example, if no 0.20 coins were available in the example of Table 2, it would be reasonable to assume that two 0.10 coins would be used instead. Algorithms for evaluating such change issues are well known. In essence, they involve working backwards from the highest 2s appropriate and available denomination and working down from that until the change is matched.

Using the example from Table 2 again, where the tender proffered was 10. 50 and the transaction value was 6.27, let us now assume that the O.SO coin received from the customer is accidentally placed in the 0.20 compartment. The result will be that the transaction appears to be 0.30 less than it should be, assuming that the 0.20 and 0.50 coins are close in weight terms which implies that the 0.50 coin will register as a 0.20 coin. T able 3 below sets out the result.

Table 3

Denomination Expected Actual Movement Movement 0.01 -1 -1 0.02 -1 -1 0.05 0.10 0.20 -1 +1 -1 = 0 0.50 +1 1.00 -4 -4 00 +1 +1 Plainly, a simple evaluation would show that a 0.50 coin is missing and a search can then be initiated to find if that coin is located in a different coin compartment of the drawer. This search is done by looking for discrepancies in other coin compartments.

As all coin compartments other than that for 0.20 coins match the expected movement, it is reasonable to conduct an evaluation on the 0.20 coin compartment to see if the missing 0.50 coin is present in that compartment. In the example shown in Table 3 it is clear that this is so, in which case a transaction report can be so annotated for future reference instead of being flagged as an error.

However, not all situations are as clear-cut as that set out in Table 3, For example, consider a transaction that results in change received from a customer of 5. 10, where :5 5.00 units of the currency in question are represented by a banknote and 0.10 is represented by a coin. The banknote is placed correctly in its compartment within the cash drawer, but the 0.10 coin also bounces into the same banknote compartment. As the coin is, say, two or three times heavier than the banknote, this will cause the banknote compartment to over-count by two or three notes, say a total of three notes s equating to 15.00 units of the currency in question. T he perceived addition of cash to the cash drawer is therefore 20.00 units (representing 15.00 units wrongly attributed to the misplaced coin and 5. 00 units correctly attributed to the correctly-placed banknote). With current technologies, this would give rise to an error of 14.90 units.

Though huge in proportion to the size of the correct transaction, such an error is lo obscure in origin because it does not relate in any obvious way to the faulty transaction. Consequently, it would be difficult if not impossible for an existing algorithm to explain the error without necessitating a manual count that is time- consuming and in itself, possibly inaccurate.

The implication of the simple examples shown in Tables 1 to 3 above is that if an item of currency is misplaced into a cash drawer compartment belonging to another denomination, then more items of that other denomination will count than would be expected. That proposition is often true, but not always. For example, a coin weighing one gram may not register as a complete coin if it falls into a compartment set up for counting five-gram coins. This is because minor variations in coin or banknote weights are expected and, in general, ignored by cash counting algorithms. For example, a cash counting algorithm may infer that a weight of say eleven grams in a compartment intended for five-gram coins represents two overweight (e.g. dirty) coins of the expected denomination rather than the alternative explanation which is two five-gram coins plus one one-gram coin.

Against this background, the invention provides a method for detecting misplacement of one or more cash items within a cash drawer having a plurality of compartments for cash items, the method comprising: recording the cash in each compartment; after a transaction, detecting a discrepancy between cash expected to be in the cash drawer and cash measured in the cash drawer; in the event of discrepancy, comparing the cash measured in each compartment after the transaction with the cash recorded in each compartment before the transaction to deduce any cash movement in each compartment; producing or calling-up a table of cash movements expected for each compartment in relation to the transaction; and matching the cash movement of each compartment against the table.

It is much preferred that the method of the invention further comprises storing the weight of cash measured in each compartment after the transaction and of cash recorded in each compartment before the transaction. Thus, the weight of cash in a compartment may be used to verify a count of cash in the compartment. Use of weight data in this way enables the invention to avoid inaccuracies arising from counting alone, for example the scenario of a one-gram coin in a compartment intended for five-gram coins as aforesaid.

Preferably, there is matching of cash movements of compartments with lower-than expected cash movements against the cash movements of compartments with higher- than-expected cash movements. In this case, if the cash movements match, it can simply be noted that the cash drawer contains the correct aggregate cash; conversely, if the cash movements do not match, further processing steps may be initiated. A non- exact match may be accepted as a match if the compared amounts are within a predetermined tolerance.

The method of the invention may employ using a difference between expected weight of cash in a compartment and actual weight of cash in a compartment to deduce the position of misplaced cash items. The expected weight may for example be deduced so from an expected number of cash items in the compartment.

To cater for situations where more than one cash item has been misplaced, the invention preferably comprises combining the weights of a plurality of cash items to match a discrepancy between expected and actual weight movements of cash in a container. For example, it is possible to store various combinations of cash items and the weights and/or values of those combinations and to look-up the stored combinations with reference to their weights and/or values. This may be achieved by consulting a look-up table containing cash items and combinations of cash items correlated with weights and/or cash values. For example, the look-up table may indicate what cash items or combinations of cash items match a measured weight difference and what value would be represented by those weight- matching items or combinations of items. The look-up table may also indicate what cash items or combinations of cash items match a measured denomination and what weight would o be represented by those denomination- matching items or combinations of items.

The method of the invention is apt to evaluate cashier performance by recording occurrences of misplaced cash items. For example, cashier performance may be evaluated with regard to the number of misplaced cash items in relation to the number of transactions or cash items processed.

Detailed optional method steps of the invention include ignoring any cash movements that match the expected cash movement for a compartment, identifying any compartments that contain less cash than indicated by the expected cash movements, JO and evaluating missing cash weight in compartments that contain less cash than indicated by the expected cash movements. That evaluation may involve determining the total weight of an overweight compartment and the expected weight of that compartment, and may further involve checking the missing weight against overweight compartments that indicate a higher than expected cash movement. The method may also comprise subtracting the expected weight of a compartment from the actual weight of a compartment to determine a weight difference, and comparing the weight difference with the missing weight. Where a match is found between the weight difference and the missing weight, this can simply be reported; alternatively where there is no match between the missing weight and the weight difference, this so mismatch may be accepted and reported as an unidentified error. It is of course possible to look-up a table to identify combinations of cash items that could contribute to the weight difference.

The method of the invention may also involve subtracting the expected weight of a compartment from the actual weight of a compartment to determine a weight difference for each compartment, comparing the sum of over-weight differences in overweight compartments with the sum of underweight differences in underweight compartments and declaring a match if the sums are identical or within a predetermined threshold of difference. In that event, the method may comprise summing the over-weight differences in overweight compartments and identifying a combination of cash items that matches the sum. The method may further comprise checking a matching combination of items against the sum of under-weight o differences in underweight compartments.

The invention extends to apparatus or systems adapted or programmed to perform the method of the invention.

It will be apparent that preferred implementations of the invention do not use a crude approximation of counting of actual items, but instead use the actual weight in the compartment to evaluate the potential addition of coins. Since the expected value of cash transaction is known and the weight of each compartment is known, it is possible to seek errors. In the proposed system, an error is recognised and further analysis is completed to determine the cause of the error and if needs be, automatically to correct the error.

In the implementations described, the evaluation is carried out on a central computer system. However, there is no reason why the evaluation could not be carried out at some other external location or by computer means specifically associated with the cash drawer or within a cash register having such a drawer.

Thus, the invention provides a system that seeks automatically to identify wrongly placed items in a cash drawer, preferably using weighing techniques. Put another way, so the process of the invention involves automatically identifying a denomination compartment containing items that should have been placed in a different compartment, preferably using the weight of the compartment contents to validate the identification. ll

The system preferably compares the weight of items 'missing' from one compartment of a cash drawer with the weight of 'excess' items in another compartment and deduces that the cash drawer as a whole contains the correct aggregate cash if the weights match. Similarly, the system deduces that further processing is required if the weights do not match.

The invention also provides a system, either associated with or external to the cash drawer, that makes use of the difference between the expected weight and the actual weight to deduce the positions of misplaced items, where the expected weight is lo derived from the expected number of items.

The invention also resides in a cash drawer that can determine the weight of items in the compartments as well as the value held within the compartments. These two items of information are used in parallel rather than one giving rise to the other as in known 1 5 systems.

The system of the invention evaluates the expected value of the items by weight and compares this with the actual weight to ascertain if a cash drawer compartment holds the correct amount of the items in question or if it contains more or less weight than expected. Preferably, the system combines the weights of one or more item types to assist in the identification of misplaced items. More generally, the system can combine the weights of one or more item types and store the combinations of items to case future processing.

Some embodiments of the system combine the weights of one or more item types and compare the total weight of the combination with known under- or over-weights of compartments within the cash drawer to identify combinations of cash items that may be misplaced.

Preferred embodiments of the invention store known combinations of item weights in a table to increase processing speed.

The system of the invention can evaluate cashier performance based on the number of misplaced items, for example by evaluating cashier performance as an error percentage or other ratio relating to the number of transactions or number of items processed.

The invention involves evaluating which denomination compartments are likely to have different contents after a transaction than they had before the transaction; evaluating denomination compartments whose contents have changed in this way; eliminating those compartments whose cash changes, if any, match an expected value; identifying denomination compartments that contain a wrong value; seeking to pair denomination compartments where one denomination compartment contains more l o than expected and another denomination compartment that contains less than expected; and preferably validating any pairing by using the known weight of the missing cash items and the measured weight of the denomination compartments that contain more cash than expected.

Preferred embodiments of the invention may involve determining that the number of errors in the denomination compartments exceeds a limit that may be acceptably matched, where the limit may be defined as a number of denominations that contain errors, a total number of errors, or a total value of errors. The system may, for example, define the maximum error that can automatically be corrected, specified by cash value or number of items or number of denominations. The invention may involve creating a warning to indicate that such a correction has taken place automatically. For example an icon may be displayed on a computer screen adjacent to the transaction record to indicate that such matching has taken place, or some other, independent reporting system may report on instances of matching. More generally, a warning may be created to the effect that the system could not (or would not) attempt to match the errors.

It is accepted that the techniques offered by the invention are not foolproof in all circumstances. However, the invention is intended to offer an improvement over the so existing practice of ignoring such problems. Naturally, if too many compartments contain errors it will not reasonably be possible accurately to deduce where cash has been wrongly deposited. A system for limiting the applicability or scope of this system is therefore advantageous. For example, the limit may simply be based on the maximum number of compartments that are allowed to contain errors before the process is activated. It may also be considered prudent to limit the value that can be corrected in such a manner, for such behaviour may be undertaken deliberately in order to camouflage other fraud or theft.

In order that the invention may be more readily understood, some embodiments in accordance therewith will now be described, by way of example, with reference to the accompanying diagrams in which: Figure] exemplifies a typical hardware system for implementing the method lo of the invention; Figure 2 is a flow diagram of a first embodiment of the invention when implemented on the hardware system shown in Figure 1; Figure 3 is a detailed flow diagram that further particularizes some method steps of the diagram of Figure 2; Figure 4 is a flow diagram of a further embodiment of the invention employing look-up tables; and Figure 5 is a flow diagram of a simplified embodiment of the invention.

In Figure l, there is provided weighing means 2 which may comprise the cash drawer of an intelligent cash register or some type of weighing scale with an electronic output indicative of the weight measured. Weighing means 2 is connected in turn to a processing unit 4 so that data relating to the weight of a sample placed upon the weighing means 2 may be transmitted electronically to the processing unit 4. Also connected to the processing unit 4 is a data store 6 and the processing unit 4 may read and write data to/from the data store 6. A video display unit 8, or some other means of alerting an operator or linked master control unit (not shown) to an erroneous situation, is optionally connected to the processing unit 4. For example, the processing unit 4 may, instead of being a stand-alone unit, be connected as part of a larger network (not shown in Figure 1) and receive and transmit data to a network controller (also not shown).

In use, to determine the value or number of a plurality of cash items placed on the weighing means 2, the weight reading output from the weighing means 2 is fed into the processing unit 4 which determines the cash value by processing the weight s reading output using stored values of standard weights for cash items, retrieved from the data store 6. The standard weights of each cash type likely to be encountered stored in the data store 6 may be the official weights accepted as standard values for those cash types, e.g. as published by the banknote printer or by national banking authorities.

Optionally, all of the weight readings determined by the weighing means 2 may be saved to the data store 6 for retrieval at the end of the day or working shift, together with timestamps relating to the recordal of each item.

The Applicant's co-pending European patent application EP 0724242, whose contents are incorporated herein by reference, gives an exemplary hardware context for the present invention. In this case, a multi- compartment cash drawer has weighing means associated with each compartment and the compartments may each be defined by separate containers such as coin cups. This is the context in which the following so description is to be understood, initially with reference to Figures 2 and 3 of the drawings.

For simplicity, it can be assumed that factors such as tare weights (e.g. coin container or cup weights) and other potential causes of inaccuracy such as base calibration, system drifts and environmental considerations have already been allowed for in the measuring systems. Such techniques form no part of the present invention and are generally known in the artor subject of different patent applications filed by the Applicant.

With specific reference to Figure 2, once the cash drawer of a cash register is closed, or once a transaction has otherwise been completed, the amount of cash it contains is counted at l0 by weighing the contents of each cash compartment of the drawer and correlating that weight with the standard weight values per unit, stored at 6 in Figure 1. The difference in cash between the current contents and the previous contents is is thereby evaluated and then compared with the expected difference value, which is usually obtained from a point of sales (POS) system in some manner.

After starting at step 10, a decision step 12 looks for any discrepancy between actual and expected cash differences. If the actual cash difference matches the expected cash difference, no further action is necessary and the process completes at 14. If the values do not match, a check is made to determine if the cause of the mismatch is a misplaced cash item. The first stage of this process at 16 is to prepare or call up a list or table of the expected cash movements based on the tender offered and the lo transaction value, in the manner previously described with reference to Table 2. The system will then, at 18, match the expected cash movements against the actual cash movements and, at 20, ignore any denominations that match or, at this stage, indicate a higher value than expected. Together, these steps therefore identify at 22 the denominations that are missing, i.e. compartments within the cash drawer that contain less cash than expected.

Following this, the method of this embodiment evaluates the missing weight at 24 and checks the missing weight against over-weight compartments at 26. These steps 24 and 26 are shown in overview in Figure 2 and will be described below in more detail below with reference to Figure 3. Where decision step 28 determines a match between missing weight and overweight compartments, the effects of the match are stored at and a transaction report is updated at 32. A further decision step at 34 determines whether there are any more overweight compartments; this decision step 34 is also reached if no match between missing weight and overweight compartments is found at decision step 28. If there are any more overweight compartments, the process loops back to step 26 of checking the missing weight against over-weight compartments.

Conversely, if there are no more overweight compartments, the mismatch is accepted as an unidentified error and a report is generated at 36.

The process then completes at 14 as aforesaid and will effectively detect one or more items that have been wrongly placed. If no match is found once the process has been completed, it is normally assumed that the cause is some form of cashier error or a deliberate fraud. Any reports or indicators will be annotated accordingly. An error may also be caused by complex incorrect placements of coins and notes, but this is considered to be a cashier error.

Figure 3 further particularizes steps 24 and 26 of Figure 2, namely evaluating the missing weight at 24 and checking the missing weight against over-weight compartments at 26. In other words, once a denomination compartment that contains less than the expected amount of cash has been identified, the missing weight is evaluated rather than the value of the denomination. This evaluation at steps 38 and is as follows: lo Missing weight = (expected number of cash items - actual number of cash tems)*calibration weight of item where the calibration weight is the weight known or assumed to represent one of the cash items.

At step 42, the method obtains the total weight of an overweight compartment and then determines the expected weight of the compartment at 44. The determination of expected weight is carried out in a manner similar to the determination of missing weight. In this case the expected weight is the product of the number of expected items (evaluated by considering the movement of cash items as already described) and the known calibration weight. At 46, the expected weight, determined at 44, is subtracted from the actual weight, determined at 42. Decision step 28 (as in Figure 1) compares the weight difference determined at 46 with the missing weight determined at 40. If a match is found, that match is reported at 32 and the process completes at 14, again as in Figure 1. If no match is found, the process is repeated for further overweight compartments by virtue of decision step 34 looping back to step 42 until no further overweight compartments remain whereupon the process completes at 14.

It will be apparent that the method steps of Figure 3 check the missing weight against each of the overweight compartments, regardless of whether or not their contents appear to be correct in terms of denomination count. This is to allow for the effect of a light item in a compartment calibrated for use with heavier items in which, for the reasons given previously, the light item may not register as a unit at all. Yet, in the method of the invention, the weight of even a light item will register and the resulting discrepancy between expected and actual weight of a compartment can be compared with the missing weight to determine a match.

s The embodiment of Figures 2 and 3 makes a complete evaluation of the weight information for every compartment of a cash drawer. The further embodiment of the invention shown in Figure 4 of the drawings, shows how it is possible to enhance the process such that if appropriate matches are not achieved, combinations of items could be identified.

The improvement in Figure 4 employs look-up tables to ease computational requirements. Specifically, a look-up table is created containing appropriate combinations of items. This table can be made as large as is appropriate to contain all expected (or even all known) combinations of cash items. Typically, but not necessarily, such a table will be sorted into weight order such that the weights of various combinations of items (based upon known calibration weights for each item) are in order. As an example, Table 4 below is a table that allows reverse look-up, in effect inferring what items or combinations of items match a measured weight and what value would be represented by those weight-matching items or combinations of items.

In Table 4, the numbers in the four central columns identify the number of items of each denomination that match the weight given in the first column. The final column contains the matching value. It could be equally appropriate to sort the table by this value rather than by the weight. In practice, there would be many more rows and columns in the table, representing most if not all combinations likely to be encountered in reality.

Table 4

Denomination: 0 01 0.02 0.05 0.10 Calibration Weight: 1.5 grams 2 grams 4 grams 7 grams Weight/ Denomination 0.01 0.02 0.05 0.10 Value 4 grams 2 0.04 4 grams 1 0.05 4.5 grams 3 0.03 5.5 grams 1 1 0.06 6 grams 1 1 0.07 6.5 grams 1 1 1 0.08 7 grams 1 0.10 7 grams 2 1 0. 12 It should be noted that some items or combinations of items may be ambiguous in the lo sense that a given weight may indicate more than one possible item or combination of items, which may also have a different value or aggregate value. For example, the first two rows of the table show how a weight of 4 grams may indicate that there are two 0.02 coins each weighing 2 grams or one 0.05 coin weighing 4 grams. Obviously in the former case, the aggregate value is 0.04 whereas in the latter case, the value is 0.05.

The process of identifying missing items will closely follow the processes described above for the embodiment of Figures 2 and 3. Consequently, like numerals are used for corresponding method steps. In essence, Figure 4 corresponds to Figure 2 save that method steps 20 to 26 are replaced by new steps 48 to 54. Briefly, instead of seeking a direct match when evaluation of overweight compartments occurs, a look-up operation is performed upon the table to identify combinations of items that could contribute to such a weight. These possibilities are then checked against the known discrepancy and any matches are reported or otherwise indicated.

With reference to Figure 4 in comparison with Figure 2, step 48 corresponds to step s 20 in that it eliminates any movements that match but it does not eliminate any movements that indicate a higher value than expected. Instead, and in contrast to step 22 of Figure 2, step 50 identifies compartments within the cash drawer that contain more weight than expected. Step 52 then evaluates the additional weight of one of those compartments and step 54 refers to the look-up table to identify the value of lo items that might match the weight difference determined at 52. The result of decision step 28 depends upon whether or not a match is found between the value determined at 54 and the weight difference determined at 52. After decision step 28, the method corresponds to that shown in Figure 2, save that decision step 36 loops back to step 52 rather than step 26 in the event that there are any more overweight compartments to assess.

In the simpler (but not currently preferred) embodiment of Figure 5, the sum of the overweight compartments is compared with the sum of the weights of the missing items and if the match is sufficiently close a balance would be declared. This has the JO disadvantage that reporting cannot be made in detail, but conversely has the advantage of simple processing. It may be beneficial to use such an algorithm as an easy check to determine whether or not additional processing is necessary.

Looking in detail at Figure 5, this broadly corresponds to Figure 4 so again, like as numerals are used for corresponding method steps. Here, the main differences are new steps 56 and 58 in which, respectively, all of the overweight compartments are summed and all of the underweight compartments are summed. If a match is found between the sums determined at 56 and 58, the result is stored at 30, reported at 32 and recorded at 36 as an (as yet) unidentified error. If no match is found, the result is also recorded at 36 but without the intermediate storage and reporting steps 30 and 32.

Unlike the embodiments of Figures 2 and 4, there is no need for iteration that seeks further overweight compartments and loops back for further processing.

Many variations are possible within the inventive concept. For example, the embodiment of Figure 4 is but one of many implementations of the invention where some form of combinatorial data is used. Another example of a suitable process would be to identify the missing items by denomination and work back seeking the weight difference. In general, the invention may involve seeking underweight compartments having found one or more overweight compartments, or seeking overweight compartments having found one or more underweight compartments.

A further implementation would be to sum all the over-weights from the various lo compartments and then use that to identify a combination of items. This combination of items would then be checked against the missing items and again a match would result in a report or other form of indication.

It should also be noted that whilst the above process is exemplified in the context of a cash register drawer, the method of the invention is equally applicable to all cash counting machines using weighing techniques, covered or not, indoor or not. For example, this process is equally applicable to the open hod-type weighing machines in use today.

Claims (32)

1. I. A method for detecting misplacement of one or more cash items within a cash drawer having a plurality of compartments for cash items, the method comprising: recording the cash in each compartment; after a transaction, detecting a discrepancy between cash expected to be in the cash drawer and cash measured in the cash drawer; in the event of discrepancy, comparing the cash measured in each compartment after the transaction with the cash recorded in each compartment before the transaction to deduce any cash movement in each compartment; producing or calling-up a table of cash movements expected for each compartment in relation to the transaction; and matching the cash movement of each compartment against the table.
2. 2. The method of Claim 1, comprising storing the weight of cash measured in each compartment after the transaction and of cash recorded in each compartment before the transaction.
3. 3. The method of Claim 1 or Claim 2, comprising matching the cash movements of compartments with lower-than-expected cash movements against the cash movements of compartments with higher-than-expected cash movements.
4. 4. The method of Claim 3, comprising noting that the cash drawer contains the correct aggregate cash if the cash movements match.
5. 5. The method of Claim 3, comprising initiating further processing if the cash movements do not match.
6. 6. The method of any of Claims 3 to 5, wherein a non-exact match is accepted as a match if the compared amounts are within a predetermined tolerance.
7. 7. The method of any preceding Claim and using a difference between expected weight of cash in a compartment and actual weight of cash in a compartment to deduce the position of misplaced cash items.
8. 8. The method of Claim 7, wherein the expected weight is deduced from an expected number of cash items in the compartment.
9. 9. The method of any preceding claim, wherein the weight of cash in a compartment is used to verify a count of cash in the compartment.
10. 10. The method of any preceding claim, comprising combining the weights of a Is plurality of cash items to match a discrepancy between expected and actual weight movements of cash in a container.
11. 11. The method of any preceding Claim, comprising storing various combinations of cash items and the weights and/or values of those combinations and looking-up the stored combinations with reference to their weights and/or values.
12. 12. The method of Claim I 1, comprising consulting a look-up table containing cash items and combinations of cash items correlated with weights and/or cash values.
13. 13. The method of Claim 12, wherein the look-up table indicates what cash items or combinations of cash items match a measured weight difference and what value would be represented by those weight-matching items or combinations of items.
14. 14. The method of Claim 12 or Claim 13, wherein the look-up table indicates what cash items or combinations of cash items match a measured denomination and what weight would be represented by those denominationmatching items or combinations of items.
15. 15. The method of any preceding Claim and being used to evaluate cashier performance by recording occurrences of misplaced cash items.
16. 16. T he method of Claim 15, wherein cashier performance is evaluated with regard to s the number of misplaced cash items in relation to the number of transactions or cash items processed.
17. 17. The method of any preceding Claim, comprising ignoring any cash movements that match the expected cash movement for a compartment.
18. 18. The method of any preceding Claim, comprising identifying any compartments that contain less cash than indicated by the expected cash movements.
19. 19. The method of Claim 18, comprising evaluating missing cash weight in s compartments that contain less cash than indicated by the expected cash movements.
20. 20. The method of Claim 19, comprising determining the total weight of an overweight compartment and the expected weight of that compartment.
21. 21. The method of Claim 19 or Claim 20, comprising checking the missing weight against overweight compartments that indicate a higher than expected cash movement.
22. 22. The method of Claim 21, comprising subtracting the expected weight of a 2s compartment from the actual weight of a compartment to determine a weight difference, and comparing the weight difference with the missing weight.
23. 23. The method of Claim 22, comprising reporting a match found between the weight difference and the missing weight.
24. 24. The method of Claim 22 or Claim 23, comprising accepting and reporting a mismatch as an unidentified error where there is no match between the missing weight and the weight difference.
25. 25. The method of any of Claims 22 to 24, comprising looking-up a table to identify combinations of cash items that could contribute to the weight difference.
26. 26. The method of any preceding Claim, comprising subtracting the expected weight s of a compartment from the actual weight of a compartment to determine a weight difference for each compartment, comparing the sum of over-weight differences in overweight compartments with the sum of under-weight differences in underweight compartments and declaring a match if the sums are identical or within a predetermined threshold of difference.
27. 27. The method of Claim 26 comprising summing the over-weight differences in overweight compartments and identifying a combination of cash items that matches the sum.
28. 28. The method of Claim 27, comprising checking a matching combination of items against the sum of under-weight differences in underweight compartments.
29. 29. A method of measuring cashier performance when using a cash drawer having a plurality of compartments for cash items, the method comprising: detecting a discrepancy between cash expected to be in the cash drawer and cash measured in the cash drawer; recognising the discrepancy as being due to misplacement of one or more cash items within the cash drawer; and recording such occurrences of misplacement.
30. 30. The method of Claim 29, comprising measuring cashier performance with regard to the number of misplaced cash items in relation to the number of transactions or cash items processed.
31. 31. Apparatus or systems adapted or programmed to perform the method of any preceding claim.
32. 32. A method for detecting misplacement of one or more cash items within a cash drawer, substantially as hereinbefore described with reference to any of Figures 2 to 5 of the accompanying drawings.