GB2610904A

GB2610904A - Inventory control method and system

Info

Publication number: GB2610904A
Application number: GB2209728.1A
Authority: GB
Inventors: Fenton Graham
Original assignee: Codegate Ltd
Current assignee: Codegate Ltd
Priority date: 2021-07-01
Filing date: 2022-07-01
Publication date: 2023-03-22
Anticipated expiration: 2042-07-01
Also published as: GB202109556D0; GB202209728D0; GB2610904B

Abstract

Stock control system 100 where the number of items 125 in each of a plurality of receptacles 110, 115, 120 is determined using weight measurement of receptacles and tare calibration. Each bin or basket has a load sensor comprising a strain sensor which outputs a weight value. The weight parameter is transmitted along with a unique ID via a radio frequency transmitter to a reader 130. The reader receives the rf signal and a processor identifies a number of items stored in each of the bins based on the weight measurement parameter and at least one calibration value (e.g. tare). Preferably, the at least one reader comprises a gateway receiver linked to a remote server. Preferably, the tare calibration value is retrieved from a database. The radio frequency transmitter may have a RFID tag. The system is preferably used for tracking items that are too small or cheap to track using individual bar codes or RFID tags. Also included are inventory control methods.

Description

Inventory Control Method and System

Background

The present application relates to inventory control, particularly to inventory control of small and/or low value items.

Supply chain tracking and inventory control have become more sophisticated in recent years. Two-dimensional barcodes and radio frequency identification (RFID) tags are now used alongside conventional one-dimensional barcodes to improve visibility into supply chains. RFID tags, in particular, facilitate automated management of inventory and are now available at very low cost.

However, some items may be too small to carry an RFID tag, may have insufficient value to justify even a very low cost tag or may not have one applied at manufacture or within the supply chain for other reasons. Barcodes do not provide a solution in all scenarios because when several items are located together (such as in a parts bin) it is often impossible to read the individual barcodes.

Another drawback with providing individual barcodes or RFID tags per item is that a large database may be required to identify the unique ID numbers from a large number of identical items when only the quantity of such items is required to manage inventory.

One suggestion is to use cameras to count items but this is not always possible and often relies upon specific alignment with the items and/or the container that holds them. Many storage environments will lack adequate lighting for such arrangements to function.

It is an object of the present invention to ameliorate these drawbacks.

According to a first aspect of the present invention, there is provided a system for inventory control comprising a plurality of receptacles for each holding a plurality of items, the receptacles each comprising at least one load device having a unique identifier within the system, the load device comprising a strain sensor for deriving a weight parameter and a radio frequency transmitter for transmitting the weight parameter and the unique identifier to a reader, and at least one reader comprising a radio frequency receiver for receiving the transmitted weight parameter and the unique identifier from each of the plurality of load devices, and a processor arranged to identify a number of items held by each of the receptacles in response to the weight parameter and at least one calibration value determined from the unique identifier.

The system according to the first aspect of the present invention thus relies on measuring the weight of multiple items in one of a plurality of receptacles and transmitting an indication of the weight to a reader. The reader may then obtain calibration information relating to the weight of the receptacle (tare) and the weight of the items to derive a number of items in the receptacle. The system is thus able to convert an imprecise weight parameter into useful information. The reader may convey the number of items to a user and/or to further inventory control systems. The reader may communicate with hundreds of receptacles and, where necessary, has a plurality of distributed antennas to facilitate this. A calibration value for individual sensors may be stored remotely and may be updateable. In that way as sensors age the system can be re-calibrated, for example as part of the process of adapting the system to a new item. The ability to recalibrate as sensors age allows the use of inexpensive sensors which will wear and flex to differing degrees as they and the hardware wears. Additionally or alternatively an item calibration value will usually be stored to enable the sensor output to be converted to a quantity of items. The process may involve converting the sensor values to an approximate linear weight measurement which may be in arbitrary units or may be approximately calibrated in known weight units and then applying an item value or may involve converting sensor values directly to item counts without using intermediate units.

The precision of the load device can be low and hence they can be manufactured cheaply. This is because the system does not need to identify the precise weight of any particular item but only distinguish between integer numbers of items. In certain applications, even per-item precision is not required. For example, where a large number of a particular item are stocked and must be re-ordered if the quantity falls below 100, it generally does not matter if the re-stocking is triggered at 100 units, 105 units or even 95 units. Moreover, although a low level of absolute accuracy may be unacceptable if trying to get an accurate weight measurement to use, when counting in items and appropriately calibrated, the accuracy may well reduce to a relatively accurate item count.

In one embodiment, the transmitter is preferably a Bluetooth TM transmitter. In order to provide energy for the strain sensor and the transmitter, the receptacles include an electrical power source such as a solar panel.

In another embodiment, the receptacles preferably do not require an electrical power source so that the strain sensor and radio frequency transmitter receive electrical power derived from an external radio frequency signal. The radio frequency transmitter preferably comprises a RFID tag.

Where the radio frequency transmitter is a RFID tag, the receptacles may be provided with a battery so that the RFID tag can operate in a battery-assisted-passive (BAP) mode that has increased operating 30 range.

Some strain sensors are sensitive to temperature variations in which case the load device of a receptacle preferably further comprises a temperature sensor coupled to the radio frequency transmitter. This may be incorporated into the RFID tag.

The load device may be mounted in any position relative to the receptacle provided that the strain sensor is deflected when items are loaded onto the receptacle. In one arrangement the load device is located substantially below the receptacle while in another it is located substantially to one side of the receptacle.

While a wide variety of receptacles may be used, one convenient form-factor comprises a bin.

The reader of the system preferably comprises a barcode reader to facilitate identification of a receptacle by an operator. The reader is preferably arranged to alert an operator when a number of items held by a particular receptacle is at or below a threshold. This alert may be via a user interface (UI) to an operator or may be via an automated message to another system responsible to re-ordering etc. The reader may be arranged to operate automatically, reading each of the receptacles periodically.

The reader may be arranged to read a temperature parameter from a load device which may then be applied to a temperature compensation calculation. Alternatively, or in addition, a temperature reading may be read to ensure that temperature-sensitive goods have not been subject to extremely low or high temperatures.

According to a second aspect of the present invention there is a provided a method of calibrating a receptacle, the method comprising identifying a first number of items held by the receptacle, receiving a first weight parameter and a unique identifier from a load device corresponding to the receptacle, identifying a second number of items, different from the first, held by the receptacle, reading a second weight parameter from the load device corresponding to the receptacle, and storing, in association with the unique identifier, at least two calibration values based on the first and second weight parameters.

To improve the accuracy of the inventory system, the individual load devices may be calibrated such as by an operator of a reader. The calibration parameters may then be stored in a database and consulted when the corresponding receptacle is to be read. Preferably two calibration values are stored -the weight parameter from the load device when the receptacle is empty (i.e. tare) and the change in the weight parameter when a single item is added or removed (per-item calibration value).

The reader may conveniently instruct the operator to remove or add items as appropriate while the weight parameters are read and then perform the calculation to derive the per-item calibration value.

A barcode mounted on the receptacle can be read by the reader to ensure that the operator is using the correct receptacle.

Where necessary (for example where large temperature variations may be expected), the reader may also be arranged to read and store a temperature parameter from the load device.

In some systems the calibration process may not be required as the calibration values (tare and per-item) may be inferred from other information. This can be done if the tare of the receptacles and the weight of the items is known (for example from inventory database(s)). For instance, assume that the receptacles typically weigh 100g, the item is a bolt that typically weighs 23g and the full range of the load device is 3Kg. If the output of the load device is 80% of maximum (equating to approximately 2.4Kg) then it can be inferred that there are 100 bolts in the receptacle ((2400-100)/23) = 100). In many applications it will not matter if the number of bolts is wrong by a small amount or that the system believes the receptacle to be empty despite it still containing a few bolts. Moreover, if the sensor typically has an accuracy of about 10%, that may mean a weight reading is 30 grams out which may be unacceptable as a weighing device. However, that 30g is only an error of a count of one item.

Further, with calibration measuring the reading at a full (or near full) integer count of actual items and an empty (or other near empty) count and working in units of items it may be found that actual count errors are small.

Even without accurate values for the tare and per-item calibration values, the system will still function satisfactorily. It is helpful but not critical that the output of the strain sensors and the weight of the items is consistent. Furthermore, if the individual items are relatively heavy even lower accuracy can be tolerated without impacting the utility of the system.

According to a third aspect of the present invention there is provided a method of determining the contents of a receptacle, comprising identifying a receptacle to be read, obtaining at least one calibration value for the receptacle, receiving a weight parameter and a unique identifier from a load device associated with the receptacle, deriving a number of items borne by the receptacle from the weight parameter and the at least one calibration value, and storing the number of items in association with the unique identifier.

The number of items is preferably conveyed to an operator, for example via a display. Where a number of items has fallen below a re-stocking threshold, the reader may warn an operator by means of an alarm. Equally, or additionally, the reader may inform another computer system to re-stock or re-order the necessary items.

In common with the second aspect, the reader may be provided with a barcode reader that is used to ensure that the correct receptacle is being interrogated.

According to a fourth aspect of the present invention, there is provided a method of determining the contents of a plurality of receptacles, comprising:, receiving a weight parameter and a unique identifier from a load device associated with a plurality of receptacles, and, for each receptacle: obtaining at least one calibration value for the receptacle deriving a number of items borne by the receptacle from the weight parameter and the at least one calibration value, and storing the number of items in association with the unique identifier.

The fourth aspect of the invention provides a complete or continuous stock-taking function. While reading the contents of receptacles in a similar manner to the third aspect, this aspect automates the process to periodically receive data from each receptacle in turn. The fourth aspect may comprise identifying a plurality of receptacles to be read.

Preferably, the method further comprises a obtaining a minimum stock parameter for the particular receptacle, comparing the number of items borne by the receptacle with the minimum stock parameter, and alerting an operator if the minimum stock parameter has been satisfied. Alternatively a re-stocking command may be communicated to another computer system.

The methods of the second, third and fourth aspects are preferably executed by apparatus comprising a radio frequency receiver and a processor adapted to perform one or more of the methods.

The apparatus may include a barcode reader and comprise a hand-held device.

The methods of the second, third and fourth aspects of the invention may be carried out by a program comprising instruction executed on a computer.

According to a fifth aspect of the present invention, there is provided a receptacle for storing a plurality of items, the receptacle comprising: a first portion and a second portion which is movable relative to the first portion in response to the receptacle carrying stored items, and a load device, the load device comprising: a substrate arranged to deflect in response to the relative movement of the first and second portions, the substrate comprising: a strain sensor responsive to deflection of the substrate to provide a deflection signal, a radio transmitter for transmitting a unique identifier and a parameter in response to the deflection signal and an antenna coupled to the radio transmitter.

In one embodiment, the radio transmitter is preferably a Bluetooth TM transmitter. In order to provide energy for the strain sensor and the transmitter, the receptacles include an electrical power source such as a solar panel.

In another embodiment,the radio transmitter is preferably a RFID tag which allows the receptacle to be a passive device. The RFID tag preferably further comprises an analogue-to-digital converter for digitizing the deflection signal. While a UHF tag is preferred, other types of RFID tag may be used.

Since strain sensors can be affected by fluctuations in temperature, the radio frequency identification tag preferably further comprises a temperature sensor to permit temperature corrections to be made.

While it is preferred that the receptacle is a passive device, provision of electrical power, such as a battery, will allow greater transmission range.

The substrate is preferably a printed circuit board (PCB) which carries the strain sensor, radio frequency transmitter and antenna. This means that the load device can be manufactured at very low cost.

The location of the load device in the receptacle is flexible, provided that when a load is applied to the receptacle, the substrate deflects. For example, the load device may be located substantially below or behind the receptacle. A common receptacle with many applications is a storage bin although many form factors are suitable. On example is a shelf such as would be found in a retail store.

According to a sixth aspect of the present invention there is provided load device for use in a receptacle for storing a plurality of items, the receptacle comprising a first portion and a second portion which is movable relative to the first portion in response to the receptacle carrying stored items, the weight sensor comprising: a substrate adapted to mount to the first and second portions and arranged to deflect in response to relative movement of the first and second portions, the substrate comprising: a strain sensor responsive to deflection of the substrate to provide a deflection signal, a radio transmitter for transmitting a unique identifier and a parameter in response to the deflection signal and an antenna coupled to the radio transmitter.

The radio transmitter is preferably a RFID tag which allows the receptacle to be a passive device. The RFID tag preferably further comprises an analogue-to-digital converter for digitizing the deflection signal. While a UHF tag is preferred, other types of RFID tag may be used.

Since strain sensors can be affected by fluctuations in temperature, the radio frequency identification tag preferably further comprises a temperature sensor to permit temperature corrections to be made. Alternatively, or in addition, a temperature reading may be taken to ensure that temperature-sensitive goods have not been subject to extremely low or high temperatures.

The present invention will now be described, by way of example, with reference to the accompanying drawings, in which: Figure 1 shows a schematic diagram of an inventory system according to an embodiment of the invention, Figure 2 shows a cross-section through an exemplary parts bin according to an embodiment of the invention, Figures 3a and 3b show an exemplary wall-mounted parts bin according to an embodiment of the invention, Figure 4 shows a perspective view of an alternative receptacle according to an embodiment of the invention, Figure 5 is an exploded diagram of the receptacle shown in Figure 4, Figure 6 shows an alternative, larger, receptacle according to another embodiment of the invention, Figure 7 shows a block schematic diagram of a reader and load devices according to an embodiment of the invention, Figure 8 shows a portion of a database accessed by the reader of Figure 7, Figure 9 shows a flow diagram for calibration of receptacles according to an embodiment of the present invention, Figure 10 shows a flow diagram for an on-demand read of inventory according to an embodiment of the present invention, Figure 11 shows a flow diagram for a continual read of inventory according to an embodiment of the present invention, Figure 12 shows a diagram of a load device according to an embodiment of the invention, Figure 13 shows a diagram of a load device according to an embodiment of the invention Figure 14 shows a schematic diagram of another sytem according to an embodiment of the invention Figure 15 shows a block diagram of another receptacle according to an embodiment of the invention, and Figure 16 shows a block system digital diagram of a system according to an embodiment of the invention.

Detailed Description

Figure 1 shows a system 100 comprising a plurality of storage bins 110, 115, 120 and a RFID reader 130. The reader 130 includes a user interface (UI) comprising a plurality of push buttons 140 and a LCD screen 135. Other arrangements may be alternatively or additionally provided such as a touch screen, voice-response system, audio alerts and so on.

Each of the bins 110, 115, 120 includes a load device (described below) for detecting the presence of items 125 in the bins and for communicating wirelessly with the reader 130. The wireless interface is provided by ultra high frequency (UHF) radio frequency identification (RFID) tags. When interrogated by the reader, the RFID tags wake up, provide power to a strain sensor, receive a weight parameter from the sensor and transmit this back to the reader together with the unique ID of the RFID tag. The reader then accesses a database entry for the particular bin to which the tag corresponds and updates it as appropriate.

The bins are preferably also provided with a unique barcode such as a 2-dimensional barcode to permit identification by an operator during configuration and stock-taking. These operations are described further below with reference to figures 9 and 10.

Although the embodiment comprises bins carrying a purely passive load device, the bins may be provided with batteries to increase the radio range of the RFID tags in a battery-assisted passive (BAP) mode.

Figure 2 shows a cross-section through an exemplary parts bin 200. The bin may be supported by a horizontal surface such as a shelf or wall-mounted as desired. The bin comprises an outer body 205 similar to known parts bins as well as a hinged platform 210 which substantially covers the floor of the bin. The platform 210 is connected to the rear wall of the bin by a hinge 215 which allows it rotate slightly within the bin. Towards the front of the bin the platform 210 is supported by a load device (combined RFID tag and strain sensor) 220 arranged between the platform and the floor of the bin. As the platform rotates downwards around the hinge (in response to the weight of items in the bin) the load device is subject to strain which allows measurement of the weight of items in the bin. An embodiment of the load device is described below but other arrangements of load cells and communication devices will be apparent to the skilled reader.

Figures 3a and 3b show a wall-mounted bin 300 having a load device 315 arranged between the bin and a wall 305. Figure 3a shows a side view in which the bin is supported along its top rear edge by a hinge 310 which allows the bin to swing slightly away from and towards the wall. Located near to the bottom of the rear face of the bin, and preferably equally spaced between the sides of the bin is the load device 315. As items are placed in the bin, it rotates slightly around the hinge 310 to apply strain to the load device 315 and allow measurement of the weight of the items.

Figure 3b shows a larger scale plan view of the bin 300 showing two clips 310 and 315 which may be provided as mouldings on the bin. The clips each provide a groove, closed at its lower end, substantially parallel to the rear wall of the bin. The load device, which comprises a rectangular substrate 330, is slid downwards into the two grooves. The load device also has a protrusion 335 which touches the wall 305 when the bin is mounted. As items are placed in the bin, it rotates slightly around the hinge to apply stress to the substrate. This is measured by a stress sensor which is converted into a weight parameter for transmission by the RFID tag.

Figure 4 shows an alternative receptable according to an aspect of the present invention. The figure shows a receptacle 400 carrying a conventional parts bin 405. The receptacle comprises a base 410 and a top plate 415 and is also provided with a 2-dimensional bar code 420 used for calibration and targeted stock enquiries. This receptacle is more suited to items that stack or to support a container that, in turn, holds items. In an embodiment the receptacle is 160mm long, 90mm wide and 26mm high. The container may have a larger footprint than the receptacle, but for reasons of stability, should have a footprint no more than double that of the receptacle. A smaller container may be used but this will have consequences for the use of storage space. The container should have a footprint of at least 80% of the receptacle. Because the receptacle has a low height profile, it can readily be retrofitted to many warehouse and picking environments.

Figure 5 shows an exploded diagram 500 of the receptacle shown in Figure 4. The construction bears some resemblance to known scales. From the bottom upwards it comprises 4 screws 505, a base 510, 4 spacers 530, a printed circuit board (PCB) 535, 4 nuts 540, 4 compression springs 545, 2 lever arms 550 and 555, two spring closures 565, a top plate 570 and a rubber grip deck 575. The base 510 further comprises 4 upwardly-extending posts 515, 4 grooves 520 arranged in 2 pairs towards the corners of the base and 2 closure loops 525 of which only one is visible. The two lever arms 550, 555 are arranged to interlock at their mid-points to form an X-shape and the lower arm 550 includes a downwardly protruding nodule 560 at the centre point. The top plate 570 further comprises 4 lever pivots (not visible) which, when assembled, engage with the two lever arms to transfer any load borne by the top plate to the lever arms. The top plate also carries two closure loops (not visible) similar to the closure loops 525 on the base.

When assembled, the screws 505 pass through the base 510, the spacers 530 and the PCB 535 to be secured by the nuts 540. The PCB is supported by the spacers 530 near each of its four corners and is consequently able to deflect downwards in its centre. The four springs 545 are located over their respective posts 515 and the inter-locked lever arms are engaged with the four grooves 520 in the base. This locates the nodule 560 above the centre of the PCB 535. The top plate rests on the compression springs 545 and two lever arms so that the weight of the top plate and anything placed thereon is transmitted to the lever arms. The spring closures 565 are arranged between respective closure loops in the top plate and the base to prevent the parts of the receptacle from separating when the receptacle is moved from its normal orientation. The spring closures are simply intended to secure the top plate to the base and do not provide much, if any, force between the top plate and the base. The rubber grip deck 575 is glued to the top plate to improve the stability of loads placed on the receptacle. As an alternative to the rubber grip deck, a number of adhesive fixing pads, typically four, may be provided on the top plate to secure e.g. a conventional parts bin.

In operation, any weight borne by the top plate 570 will compress the springs 545, moving the top plate closer to the base. This movement is transmitted to the lever arms 550, 555 and, in turn, via the nodule 560 to cause a deflection of the PCB 535. The PCB carries the load device to measure the deflection and transmit it to the reader as described below. The base 510, top plate 570 and/or posts 515 are dimensioned to provide and end-stop that prevents the top plate from moving to such an extent relative to the base that the PCB 535 could be damaged.

Receptacles having different sensitivities (and maximum loads) can readily be provided by using stiffer or lighter compression springs 545. In one embodiment the maximum load is 3Kg. A wide range of possible loads is possible, such as 0-1kg, 0-10kg and 0-50kg. In a preferred embodiment, the receptacles are provided with excess load tolerance meaning that they will suffer no permanent damage if an excessive load is placed upon them. A typical excess load tolerance would be 25 times the maximum weighable load.

Figure 6 shows an alternative receptacle according to another aspect of the present invention, in this case a shelf 600 such as might be used in a retail store. The shelf is provided with a back-panel 605 to permit wall-mounting. Other shelving arrangements will be apparent to the skilled reader. Along the top of the back-panel is a hinge 610 allowing slight rotation in a manner analogous to that of the bin shown in Figures 3a and 3b. The shelf 600 is provided with a combined RFID tag and weight sensor 615 to permit measurement of the weight borne by the shelf.

Figure 7 shows a block schematic diagram of a system according to an embodiment of the present invention. A reader 700 comprises a processor 705 coupled to an RFID tag reader 710, a 2-dimensional barcode reader 715, a database 720, a user interface (UI) 725 and a display 730. The reader 700 communicates wirelessly with load devices 735, 740 and 745 and optically with 2-dimensional barcodes.

The reader may be a handheld device as shown in Figure 1 or may be a device or system having one or more larger displays, user interfaces and distributed antennas. Where the receptacles are larger as shown in Figure 6, more than one of the RFID tag and weight sensors may be associated with the same receptacle.

The barcode reader is optional but is useful during setup and stock control. The operation of the reader 700 as will be described with reference to Figures 9, 10 and 11 below.

As the skilled reader will appreciate, the reader functionality may be divided between different locations. For example, the reader 710 may be wirelessly coupled to a further transmitter, such as a cellular or WiFi transmitter with the remaining processing being carried out at a remote server.

Figure 8 shows a table 800 stored in the database 720 although other database arrangements will be apparent to the skilled reader. The table includes a row for each receptacle in the inventory system. The database may be located within the same housing as other components of the reader or may be located remotely and accessed, for example, by a wireless interface.

The table 800 includes fields for stock-keeping-unit (SKU) 805, product name 810, RFID tag ID 815, barcode 820, tare 825, per-item calibration 830, stock level 835, minimum stock level 840 and temperature 845. The SKU is a unique code, understood throughout the supply chain, which identifies the product named in the field 810. The tag ID 815 is the unique ID which is applied to the RFID tag and to which it responds when interrogated. The barcode is a machine-readable optical code associated with the relevant receptacle. It is used during certain reader operations as discussed further below. The tare 825 is an indication of the weight parameter when the bin is empty and the per-item calibration 830 is an indication of the change in the weight parameter when a single item is added or removed. The stock level 835 is the estimate of the number of units in the receptacle and the minimum stock level 840 is the number of units at which re-stocking or re-ordering should occur. The temperature parameter is the last known temperature of the load device (preferably taken at the time that it was last read) and can be used to allow for variations in the calibration of the load device.

Alternatively, or in addition, a temperature reading may be taken to ensure that temperature-sensitive goods have not been subject to extremely low or high temperatures.

Further fields may be added to this table as required such as further tag ID(s) in the case of multiple tags per receptacle and additional product details. Equally, not all of the illustrated fields will be required in all situations. For example, the barcode will be absent when the receptacles do not have barcodes, the SKU and barcode may be the same and minimum stock will only be required if an alarm or automated re-stocking is required when the stock drops to this level. Multiple alarm levels may also be stored.

Operation of the reader in a calibration mode, an on-demand reading mode and a continual reading mode will now be described with reference to the flow diagrams of Figures 9, 10 and 11.

Figure 9 shows a flow chart 900 for calibrating one of the receptacles in a system according to an embodiment of the invention. The object is to derive the tare 825 and per-item calibration 830 values to store in the table 800. The tare represents the output of the load device when the receptacle is empty and the per-item calibration represents the change in the output of the load device when a single item is added to the receptacle. Once the system has obtained these two parameters, it is a straightforward task to derive a number of items in the receptacle using the following equation 1: Number of items in receptacle = (output of weight sensor -tare) / per-item calibration The calibration is initiated by an operator informing the reader that a calibration is required (via the Ul). The process starts at 905 and at step 910 the barcode reader is activated. At this point the operator points the barcode reader towards the 2-dimensional barcode associated with the receptable or bin that is to be calibrated and the barcode is read. The skilled person will readily appreciate actions to be taken if no barcode is visible to the reader and these will not be described here. Instead of using the barcode, the operator may identify the bin in some other manner such as by entering the SKU or selecting the product name from a menu displayed by the reader.

The reader now informs the operator to remove all of the items from the relevant bin at step 915. At step 920 the reader uses the 2D barcode to identify the unique ID 815 associated with the bin's RFID tag from the database table 800. The reader may optionally retrieve the name of the item 810 and display this to the operator to ensure that the correct items are used in the calibration process. The reader now uses the RFID reader 710 to interrogate the RFID tag of the relevant load device.

The tag will wake up, power up the associated strain sensor, receive a weight parameter from the sensor and then transmit a message containing both its unique ID and a representation of the weight parameter. Typically this representation will be the digitized output of the strain sensor.

Optionally, the RFID tag may also inform the reader of a temperature parameter for the tag as this can affect the output of the weight sensor. The reader may then use this parameter to compensate for any temperature-induced variations in the output of the weight sensor. A strain sensor as used in one embodiment of the present invention has an inverse linear response to temperature which can readily be compensated for. As an alternative, one or more temperature sensors may be located within the environment to provide this data on the assumption that all of the receptacles in a given area will be at approximately the same temperature. The benefit of reading the individual temperature of a receptacle is that variations caused by stock at different temperatures (such as when loaded from a cold transport vehicle) can be allowed for.

The reader will now store, at step 930, a calibration value in the Tare column of the table 800 which should correspond with the output of the weight sensor when there are no items in the relevant bin.

The calibration value may comprise a digitized output of the weight sensor or may be pre-processed in some manner by the reader.

At step 935, the reader will now inform the operator to place one or more items into the receptacle and to enter the number of items via the Ul. The reader then reads the tag again at step 940 which comprises the same operations as at step 925. The reader will then process the received weight parameter by subtracting the tare calibration value 825 and dividing the result by the number of items entered at step 935.

At step 945 the reader derives the per-item calibration using the following equation 2: Per-item calibration = (weight parameter from step 940-Tare) / (number of items at step 935) As a slight variation, the reader could simply inform the operator at step 935 to place a predetermined number of items in the bin, obviating the need for the operator to enter a particular number of items into the reader.

The per-item calibration value is then stored in the table 800 at step 950 and the calibration process ends at step 955.

The calibration process described with reference to Figure 9 may not be required in all circumstances.

If the weight of the receptacles and the outputs of the strain sensors for a given load are fairly consistent then the tare value and the per-item calibration of Figure 8 may be estimated without significant sacrifice in accuracy of the system. This is particularly true when the stocked items are relatively heavy since it will be quite straightforward to distinguish between integer numbers of such units borne by the receptacle In any case the accuracy of the system is usually not critical since most operators will not care whether they have 200 items or 210 items, for example. In most applications it will be important that the system triggers a re-stock at the appropriate time and it will not matter whether the re-stock is triggered at 50 units, 48 units or even 45 units. The system can thus tolerate an error of +/-5%, +110% or even +/-20% in certain circumstances. This permits the use of very low cost strain sensors. It is preferred for the sensors to perform to an accuracy of +/-2%.

Once the system is operational and the bins contain stock items, an operator may wish to interrogate one or more of the bins to obtain an up-to-date estimate of the contents. Figure 10 shows a flow diagram of an on-demand read of a receptacle. The process starts at step 1005 in response to a request from an operator and proceeds to read a barcode at step 1010. Again, this step could be replaced by the operator informing the reader of the relevant receptacle by other means. The reader uses the barcode to retrieve the unique ID of the relevant RFID tag from the database at step 1015. The reader may also retrieve the associated Tare value and the per-item calibration at this step.

The reader then interrogates the RFID tag at step 1020 and performs the calculation using equation 1 to derive the number of items at step 1025. This step may also include temperature reading for compensation purposes or inventory tracking where appropriate. At step 1030, the number of items is displayed to the operator and at step 1035, the reader updates the database with the number of items if appropriate. The process ends at step 1040.

In addition to providing an on-demand read to an operator, the system is preferably also arranged to continually monitor the stock in all of the receptacles. This is similar to a succession of on-demand reads performed automatically by the reader. Figure 11 illustrates such a process using a flow diagram 1100.

The process starts at step 1105 and proceeds to step 1110 in which the first row in the table 800 is selected. At the following step 1115 the reader retrieves the unique ID for the RFID tag, the number of items in the associated bin, the minimum number of items and the calibration values (tare and per-item parameter) for the first item in the table. At step 1120 the reader interrogates the RFID tag which responds with the unique ID and the weight parameter as before. At step 1125 the reader performs the calculation of equation 1 as before and at step 1130 it compares the current number of items with the previously stored number 835. If these numbers are the same, the next item in the table is selected at step 1135 and the process performs step 1115 again, this time based upon the next item from the database.

If the current number of items and the stored number of items differ at step 1130, processing proceeds to step 1140 in which the new stock level is reported to an operator. At the next step 1145 the current stock level is compared with the minimum stock level. If the current stock level is above the minimum stock level then the flow proceeds to step 1135 and processes the next item in the database. If the current stock level is at or below the minimum stock level then step 1150 is executed. Step 1150 provides an alert to an operator and/or sends a re-stocking message to an automated system as appropriate and the flow proceeds to step 1135.

The load devices associated with each receptacle in the system are thus interrogated in turn which both maintains the database record and takes the appropriate action for those items whose stock levels are too low. Variations may be applied to this process. For example, the reader may be arranged to read those receptacles containing stock items with a higher turnover more frequently.

Figures 12(a) and (b) show views of a load device according to an embodiment of the present invention. The load device is mounted to the base of a receptacle as shown in Figure 5. Figure 12(a) shows the device under no load i.e. a tare condition, while Figure 12(b) shows the device under load. This load will be detected by a strain sensor shown in Figure 13.

Figure 13 shows a plan view of a load device 1300 suitable for mounting to the receptacles described above. The load device includes a printed circuit board (PCB) 1305 which supports all of the components of the device. In operation the device 1300 may be mounted as shown in Figures 2, 35, 6 and 12 to receive a strain input from a receptacle. Attached to the PCB 1305 are a strain sensor or strain gauge 1310, an RFID integrated circuit (IC) 1315 and an antenna 1320. The strain sensor is connected to the RFID IC via an analogue interface and the antenna is connected conventionally. The RFID IC preferably operates in the ultra-high-frequency (UHF) region of 860 to 960MHz.

In many respects the RFID ID 1315 is conventional but it is additionally provided with an analogue-to-digital converter (ADC) which is used here to digitize the output of the strain sensor. Such an IC is available from Asygn, Grenoble, France asygn.com under model number AS3212. The AS3212 also includes an internal temperature sensor and can be arranged to transmit the temperature together with the unique ID and the output of the ADC. When the PCB is subject to strain as shown in Figure 12, the resistance of the strain sensor 1310 varies. When the RFID IC is powered up by incoming radio waves, the ADC is also powered and a reading taken from the strain sensor. This is then digitized and transmitted by the RFID tag together with its unique ID.

An optional battery may be connected to the RFID IC to provide a battery-assisted-passive (BAP) mode of operation. In this mode the battery provides extra electrical power to the transmitter which improves the transmission range from the typical 4 metres available under passive operation.

Figure 14 shows an alternative arrangement 1400 of a system in accordance with an embodiment of the present invention. A plurality of receptacles, of which five 1402, 1404, 1406, 1408, 1410 as shown, each comprise a low energy Bluetooth TM transmitter having a maximum range of approximately 10m. Bluetooth signals transmitted by the receptacles are received at an edge gateway receiver 1412. The gateway communicates via the cloud 1414 to provide a web portal accessible from, for example, a desktop 1416 or mobile device 1418. The edge gateway may communicate, for example, via cellular, WiFi or LAN to the receiving system. The receptacles may be provided with two-dimensional barcodes for calibration purposes as previously described.

Figure 15 shows a block diagram 1500 of one of the receptacles shown in Figure 14. Each receptacle comprises a processor 1502 coupled to at least one load sensor 1504 and a Low Energy BluetoothTM transmitter 1506 having an antenna 1508. In contrast to the embodiments based on RFID communication, the circuitry is provided with its own power source, a solar panel 1510. This preferably comprises a high-efficiency solar panel which will operate in low light conditions such as those often found in a warehouse. The receptacle may also be provided with a temperature sensor as previously described.

In operation, each receptacle periodically measures the weight of items placed upon it using at least one load cell, although 4 is a typical number. The processor receives the load figure and transmits it, together with a unique receptacle identifier, to the edge gateway receiver over the low energy BluetoothTM link. Further processing of this information may be carried out as previously described. In order to preserve energy at the receptacle, it may be provided with a quiet mode. In quiet mode, the processor compares each load sensor reading with the previous reading and, if it has not changed, no transmission will take place. Typically, after 10 such occurrences, the processor will instruct the Bluetooth transmitter to make the transmission so that the server-side system knows that the receptable is still fully operational. Where the receptacle is provided with a temperature sensor, the processor will read data therefrom and transmit it over the Bluetooth TM link. Such data can be used to allow for temperature-induced changes in load cell output and/or for ensuring that temperature-sensitive goods have not been subject to extreme temperatures.

Figure 16 shows a digital system diagram 1600 showing the interactions between the various system elements. Sensor firmware 1602 in the receptacle communicates with software 1604 in the edge gateway which in turn communicates with receptacle data gateway software 1606 at a server 1608 comprising application software, web service and an SQL server communicating with data store 1610.

The server provides the web portal 1612 as before. A handheld calibration application 1614 for communicating with a handheld device 1618 also communicates with the server via handheld data gateway software 1616. The handheld device may be a mobile phone loaded with a suitable application.

The functional requirements of the software at the edge gateway receiver are as follows: Fl To listen for BLE messages and quickly filter down to those related to those originating from the receptacles.

F2 To extract weight and, optionally, temperature values and forward on to the server-based receptacle gateway (see Figure 16, 1606), adding a receptacle ID, and optional date and time stamp.

F3 Data to be received over BLE from the receptacles, data to be transmitted over WiFi or cellular network.

F4 Preferred feature -to be able to update firmware over the air using a file received from a location on the central server. It is capable of selecting one, many or all devices within range for sequential updating. This process may be arranged to be automatic in response to the file on the server having changed.

A regular 'watchdog' signal or similar may be provided by the device so that the receptacle data gateway can detect an 'offline' condition for each edge gateway.

F6 To be able to pick up data from up to 400 weight sensors operating in close proximity in an environment with min 700 Lux lighting level. Sensors to operate 'quiet mode' firmware described above.

The functional requirements of the server-based receptacle data gateway are as follows: 11 Receive data from one or multiple edge gateways 12 Store received information directly into the SQL database being used by the web portal.

13 Monitor for 'connected' edge gateways and create an alert / log entry if a disconnect is detected.

The functional requirements of the server-based handheld data gateway are as follows: Fl Receive calibration data from multiple handhelds -including at least some of: handheld user id, location, sensor ID, tare value, weight value, scale factor, temperature value, date and time.

F2 Store received information directly into the SQL database being used by the web portal The server operates in a very similar manner to that described for the earlier embodiment.

Claims

CLAIMS1. A system for inventory control comprising: a plurality of receptacles for each holding a plurality of items, the receptacles each comprising at least one load device having a unique identifier within the system, the load device comprising a strain sensor for deriving a weight parameter and a radio frequency transmitter for transmitting the weight parameter and the unique identifier to a reader, and at least one reader comprising a radio frequency receiver for receiving the transmitted weight parameter and the unique identifier from each of the plurality of load devices, and a processor arranged to identify a number of items held by each of the receptacles in response to the weight parameter and at least one calibration value determined from the unique identifier.
2. A system as claimed in claim 1, wherein the at least one reader comprises a gateway receiver linked to a remote server which houses the processor.
3. A system as claimed in claim 1 or claim 2, wherein the at least one calibration value comprises a tare value and the processor is arranged to retrieve the tare value from a database in response to the unique identifier.
4. A system as claimed in claim 1 or claim 2 or claim 3, wherein the at least one calibration value comprises a sensor calibration value and the processor is arranged to retrieve the sensor calibration value in response to the unique identifier and to process the weight parameter based on the retrieved sensor calibration value.
5. A system as claimed in any one of the claims 1 to 4, wherein the at least one calibration value comprises an item calibration value and the processor is arranged to retrieve the item calibration value in response to the unique identifier and to process the weight parameter based on the retrieved item calibration value to determine a measure of the number of items.
6. A system as claimed in any one of the claims 1 to 5, wherein the radio frequency transmitter comprises a low energy Bluetooth TM transmitter.
7. A system as claimed in any one of the claims 1 to 6, wherein the strain sensor and the radio frequency transmitter receive electrical power from a solar panel.
8. A system as claimed in any one of the claims 1 to 5" wherein at least one of the strain sensor and radio frequency transmitter receive electrical power derived from an external radio frequency signal.
9. A system as claimed in claim 8, wherein the radio frequency transmitter comprises a RFID tag and the radio frequency receiver is a RFID reader.
10. A system as claimed in claim 9, wherein the RFID tag is arranged to receive electrical power from a battery.
11. A system as claimed in any one of the preceding claims, wherein the load device of at least one receptacle further comprises a temperature sensor coupled to the radio frequency transmitter and the reader is further arranged to read a temperature parameter from at least one load device.
12. A system as claimed in any one of the preceding claims, wherein at least one receptacle comprises the load device located substantially below the receptacle.
13. A system as claimed in any one of the preceding claims, wherein at least one receptacle comprises the load device located substantially to one side of the receptacle.
14. A system as claimed in any one of the preceding claims, wherein the receptacle comprises a bin.
15. A system as claimed in any one of the preceding claims, wherein the at least one reader further comprises a barcode reader.
16. A system as claimed in any one of the preceding claims, wherein the reader is further arranged to alert an operator when a number of items held by a receptacle is at or below a threshold.
17. A system as claimed in any one of the preceding claims, wherein the reader is arranged to periodically receive a weight parameter and the unique identifier from the load device associated with each of the receptacles.
18. A method of calibrating a receptacle, the method comprising: identifying a first number of items held by the receptacle, receiving a first weight parameter and a unique identifier from a load device corresponding to the receptacle, identifying a second number of items, different from the first, held by the receptacle, reading a second weight parameter from the load device corresponding to the receptacle, and storing, in association with the unique identifier, at least two calibration values based on the first and second weight parameters.
19. A method as claimed in claim 15, comprising identifying the first number of items as zero to provide a tare calibration value for the receptacle based on the first weight parameter, and dividing the difference between the second weight parameter and the first weight parameter by the second number of items to derive a per-item calibration value for the receptacle.
20. A method as claimed in claim 15 or claim 16, further comprising reading at least one temperature parameter from the load device, and storing, in association with the unique identifier, at least one temperature calibration value based on the temperature parameter.
21. A method of determining the contents of a receptacle, comprising identifying a receptacle to be read, obtaining at least one calibration value for the receptacle, receiving a weight parameter and a unique identifier from a load device associated with the receptacle, deriving a number of items borne by the receptacle from the weight parameter and the at least one calibration value, and storing the number of items in association with the unique identifier.
22. A method as claimed in claim 21, wherein the step of identifying a receptacle to read comprises reading a barcode associated with that receptacle.
23. A method as claimed in claim 21 or claim 22, further comprising displaying the number of items to an operator.
24. A method of determining the contents of a plurality of receptacles, comprising receiving a weight parameter and a unique identifier from a load device associated with a plurality of receptacles, and, for each receptacle obtaining at least one calibration value for the receptacle, deriving a number of items borne by the receptacle from the weight parameter and the at least one calibration value, and storing the number of items in association with the unique identifier.
25. A method as claimed in claim 24, further comprising identifying a plurality of receptacles to be read.
26. A method as claimed in claim 24 or claim 25, further comprising obtaining a minimum stock parameter for the receptacle, comparing the number of items borne by the receptacle with the minimum stock parameter, and alerting an operator if the minimum stock parameter has been satisfied or communicating a re-stocking instruction to a computer system.
27. A method as claimed in any one of the claims 12210 26, wherein the at least one calibration value for the receptacle includes retrieving a tare value and a per-item value from a database.
28. An apparatus comprising a radio frequency receiver and a processor adapted to perform the method claimed in any one of the claims 18 to 27
29. An apparatus as claimed in claim 28, further comprising a bar-code reader for identifying a particular receptacle.
30. An apparatus as claimed in claim 28 or claim 29, which apparatus comprises a hand-held device.
31. A computer program comprising instructions which, when the program is executed by a computer, cause the computer to carry out the method of any of claims 18 to 27.