Weighing-enabled shelving arrangements and methods

Abstract

In a computer-implemented method of using a weighing-enabled shelving arrangement, the shelving arrangement comprises a plurality of shelves distributed vertically within a shelving bay, an upper surface of each shelf defining an x-y plane at a respective z-axis height, the x-y plane being characterized by an x-axis width and a y-axis depth. An array of weighing elements is arranged to jointly support the shelving bay and bear the weight thereof. The method comprises receiving respective streams of weight measurement data-points corresponding to the shelving bay and to products disposed on shelves therewithin, and analyzing a pattern of the weight measurement data-points to determine a z-axis height of a location of a weight-event in which a product is added to, removed from, or moved within the shelving bay.

Description

[0001] WEIGHING-ENABLED SHELVING ARRANGEMENTS AND METHODS

[0002] FIELD OF THE INVENTION

[0003] The present invention relates to weighing-enabled shelving arrangements and methods for using them in retail establishments and inventory-management locations, and in particular to determining the location of a weight event and / or identifying products involved in the weight event in multi-shelf shelving arrangements.

[0004] BACKGROUND

[0005] As automation in retail store evolves to less-friction solutions (from ‘scan&go’ to smart carts and up to autonomous store) the need for real-time, cost effective product recognition is crucial. Retailers and sellers of consumer packaged goods can benefit from real-time shelf data, such as real-time inventory and planogram compliance, and from the use of methods for weighing and tracking products on shelves in shelving bays, display refrigerators, and arrays of shelving bays. Technical solutions have been suggested for intelligent shelving arrangements that would track the weight of products on a shelf, including changes in the weight resulting from the addition of products or the removal of products. These solutions include tracking the weight of the products on each individual shelves by dedicating weighing assemblies including multiple load cells to each shelf. The requirement inherent in existing solutions to deploy large numbers of load cells makes such solutions more expensive and more complicated to implement.

[0006] There is a need for a more streamlined and less costly solution to using weighing assemblies to detect product movements such as additions to shelves, removal from shelves, and movement within shelving arrangements, to identify respective locations of such product movements, and to disambiguate the identity of the products tracked and moved. Further, there is a need for methods and systems that provide such solutions for inventory and retail locations where the products on the shelves are not homogeneous and / or are not in known or expected locations, as current solutions are inadequate for such locations.

[0007] SUMMARY

[0008] A computer-implemented method is disclosed, according to embodiments, for using a weighing-enabled shelving arrangement which comprises (i) a plurality of shelves distributed vertically within a shelving bay, an upper surface of each shelf defining an x-y plane at a respective z-axis height, the x-y plane being characterized by an x-axis width and a y-axis depth, and (ii) an array of weighing elements arranged to jointly support the shelving bay and bear the weight thereof, each weighing element comprising at least one load cell. The method comprises: (a) receiving respective streams of weight-measurement data points corresponding to the shelving bay and to products disposed on shelves therewithin; (b) analyzing the received weightmeasurement data points to determine a z-axis height of a location of a weight-event in which a product is added to, removed from, or moved within the shelving bay.

[0009] According to embodiments, a shelving arrangement comprises (a) a shelving bay including (i) a back panel and a pair of uprights associated therewith, (ii) a plurality of shelves distributed vertically within the shelving bay, each shelf supported by a pair of brackets respectively joined to the pair of uprights, an upper surface of the shelf defining a plane at a respective height, and (iii) a shelving-unit base supporting the weight of the uprights, back panel, shelves and brackets; and (b) an array of weighing elements arranged to support the shelving bay and bear the weight thereof, and configured to generate respective streams of weight-measurement data points corresponding to the shelving bay and to products disposed on shelves therewithin.

[0010] According to embodiments, a shelving arrangement comprises: (a) an array of n shelving assemblies, n being greater than or equal to one, each shelving assembly comprising at least one shelving bay and not more than two shelving bays, each shelving bay including a respective plurality of shelves distributed vertically therewithin, an upper surface of each shelf defining a x y plane at a respective z-axis height, the x-y plane being characterized by an x-axis width and a y-axis depth; (b) an array of n+1 sets of weighing elements, each set of weighing elements comprising a single elongated weighing bar comprising a plurality of load cells or a plurality of single-load-cell weighing elements each comprising a single load cell, the weighing elements arranged to support the shelving assemblies and bear the weight thereof, and configured to transmit respective streams of weight-measurement data points corresponding to the array of shelving assemblies and to products disposed on shelves therewithin, the arranging being such that each shelving assembly rests on two consecutive sets of weighing elements; and (c) a computing unit programmed to detect a change over time in values of said weight-measurement data points indicating a weight-event, and to analyze said values to determine a location of the weight-event, wherein said determining includes identifying a shelving bay and a shelf therewithin.

[0011] A computer-implemented method is disclosed, according to embodiments, for using a weighing-enabled shelving arrangement that comprises (i) a plurality of shelves distributed vertically within a shelving bay, an upper surface of each shelf defining an x-y plane at a respective z-axis height, the x-y plane being characterized by an x-axis width and a y-axis depth, and (ii) an array of weighing elements each comprising one or more load cells, the weighing elements arranged to jointly support the shelving bay and bear the weight thereof. The method comprises: (a) receiving respective streams of weight-measurement data points corresponding to the shelving bay and to products disposed on shelves therewithin; (b) responsively to a change over time in values of said weight-measurement data points, detecting an occurrence of a weight-event in which a product is added to, removed from, or moved within the shelving bay; and (c) analyzing said weight-measurement data points to identify the product.

[0012] According to embodiments, a weighing-enabled refrigerator comprises: (a) a refrigerator including one or more shelving sections distributed horizontally therein, each shelving section comprising a vertical array of shelves, an upper surface of each shelf defining an x-y plane at a respective z-axis height; (b) an array of weighing elements each comprising at least one load cell, the weighing elements arranged to support the refrigerator and bear the weight thereof, and configured to transmit respective streams of weight-measurement data points corresponding to the refrigerator and to products disposed on shelves therewithin; and (c) a computing unit programmed to (i) receive said respective streams of weight-measurement data points, (ii) responsively to a change over time in values of said weight-measurement data points, detect an occurrence of a weight-event in which a product is added to, removed from, or moved within the refrigerator, and (iii) analyze said weightmeasurement data points to identify the product.

[0013] According to embodiments, an article of manufacture is provided as a kit comprising a pair of elongated weighing bars adapted for jointly supporting and weighing a shelving bay, wherein each of the weighing bars comprises: (i) a plurality of load cells, (ii) a spaced-apart plurality of leg supports, (iii) an upper frame member adapted to receive a portion of a load-bearing member of a base of the shelving-bay, and (iv) for each of the one or more load cells, a lower frame member comprising a first portion supported by a corresponding leg support and second portion having the respective load cell fixed thereto, and a plurality of rubber shock absorbers mediating between the upper frame member and the respective lower frame member.

[0014] BRIEF DESCRIPTION OF THE DRAWINGS

[0015] The invention will now be described further, by way of example, with reference to the accompanying drawings, in which the dimensions of components and features shown in the figures are chosen for convenience and clarity of presentation and not necessarily to scale. In the drawings:

[0016] Fig. 1 A is a schematic side view of a shelving assembly comprising a shelving bay supported by a pair of weighing bars, according to embodiments of the present invention.

[0017] Fig. IB is a schematic perspective view of the shelving assembly of Fig. 1A, with non-homogeneous groups of products disposed on the shelves of the shelving bay, according to embodiments of the present invention.

[0018] Fig. 1C is a schematic side view of a shelving assembly comprising a shelving bay supported by a an array of single-load-cell weighing elements, according to embodiments of the present invention.

[0019] Fig. ID is a schematic perspective view of the shelving assembly of Fig. 1C, with non-homogeneous groups of products disposed on the shelves of the shelving bay, according to embodiments of the present invention.

[0020] Figs. 2A and 2B are respective schematic side and end views of an exemplary weighing bar, according to embodiments of the present invention.

[0021] Fig. 2C is a schematic end view of another exemplary weighing bar, according to embodiments of the present invention.

[0022] Fig. 2D is a schematic side view of the weighing bar of Figs. 2A and 2B, with a side panel removed to show an interior volume including two load cells, according to embodiments of the present invention.

[0023] Fig. 2E is a schematic side view of the weighing bar of Figs. 2A and 2B, with a side panel removed to show an interior volume including a single load cell, according to embodiments of the present invention.

[0024] Figs. 3A, 3B, 3C and 3D illustrate exemplary single-load-cell weighing elements, according to embodiments of the present invention.

[0025] Fig. 4A shows an exemplary bending-beam load cell, according to embodiments of the present invention. Fig. 4B shows an exemplary planar load cell, according to embodiments of the present invention.

[0026] Fig. 5 is a schematic block diagram of electronic circuitry for a shelving assembly showing selected components and communications arrangements, according to embodiments of the present invention.

[0027] Figs. 6A, 6B and 9 show flow charts of methods and method steps for using a weighing-enabled shelving arrangement, according to embodiments of the present invention.

[0028] Figs. 7 and 8 show respective graphs of streams of weight-measurement data points, according to embodiments of the present invention.

[0029] Fig. 10A is a schematic perspective view of a shelving assembly supported by a pair of weighing bars and comprising two shelving bays arranged back-to-back, according to embodiments of the present invention.

[0030] Fig. 1 OB is a schematic side view of a weighing bar of the shelving assembly of Fig. 10A, according to embodiments of the present invention.

[0031] Fig. 11 is a schematic perspective view of an exemplary shelving arrangement comprising an array of three shelving assemblies, where each respective shelving assembly comprises a single shelving bay, according to embodiments of the present invention.

[0032] Fig. 12 is a schematic perspective view of an exemplary shelving arrangement comprising an array of three shelving assemblies, where each respective shelving assembly comprises two shelving bays, according to embodiments of the present invention.

[0033] Fig. 13 is a schematic front view of a weighing enabled retail refrigerator according to embodiments of the present invention.

[0034] Figs. 14A and 14B are schematic illustrations of respective weighing-enabled display refrigerators according to embodiments of the present invention.

[0035] DETAILED DESCRIPTION OF THE ILLUSTRATED EMBODIMENTS

[0036] The invention is herein described, by way of example only, with reference to the accompanying drawings. With specific reference now to the drawings in detail, it is stressed that the particulars shown are by way of example and for purposes of illustrative discussion of the preferred embodiments of the present invention only, and are presented in the cause of providing what is believed to be the most useful and readily understood description of the principles and conceptual aspects of the invention. In this regard, no attempt is made to show structural details of the invention in more detail than is necessary for a fundamental understanding of the invention, the description taken with the drawings making apparent to those skilled in the art how the several forms of the invention may be embodied in practice. Throughout the drawings, like-referenced characters are generally used to designate like elements. Subscripted reference characters (e.g., 1 Oi) may be used to designate multiple separate appearances of elements of a single species, whether in a drawing or not; for example: 10i is a single appearance (out of a plurality of appearances) of element 10. Similarly, 10IA and IOIB can be two different appearances of a more general element that can be labeled 1 Oi, and IOIAI and 101A2 can be two different appearances of a more general element labelable 10IA. The same elements can alternatively be referred to without subscript (e.g., 10 and not 10i, or 1 Oi and not 10IA, etc.) when not referring to a specific one of the multiple separate appearances, i.e., to the species in general.

[0037] Some of the terms are used within have been selected to have specific meanings for the sake of clarity and consistency. The term ‘shelving bay’ as used herein means a structure comprising one or more shelves displaced vertically from each other, arrayed on a single side of a back panel of the shelving bay. Figs. 1 A and IB illustrate an example of a shelving bay. A shelving bay can have two vertical elements known in the industry as ‘uprights’ which have various mechanical arrangements for hanging shelves. When shelving bays are aligned laterally in close proximity to each other, the design of the uprights may enable each upright to be shared by the respective adjacent pair of shelving bays.

[0038] A ‘shelving assembly’ as used herein means an assembly that comprises a single shelving bay or a pair of back-to-back shelving bays.

[0039] A ‘shelving arrangement’ as used herein means a collection of shelving assemblies of any type and quantity.

[0040] A ‘weighing element’ as used herein means a structure or module comprising one or more load cells and designed for supporting, in combination with one or more other weighing elements, a shelving arrangement. A weighing element according to embodiments can be an elongated weighing bar comprising one or more load cells and designed for placement under and support of one side of a shelving bay, or a single-load-cell weighing element not designed to be the only weighing element per under one side of the shelving bay, as shall become clearer in the illustrated embodiments disclosed hereinbelow.

[0041] Some embodiments of the disclosure relate to shelving arrangements adapted for use with weighing elements for purposes of tracking and / or controlling inventory and / or retail activities corresponding to products stored or displayed within the shelving arrangements.

[0042] The shelving arrangements can include various kinds of shelving bays arranged in various combinations, and in some embodiments include refrigerators and coolers, e.g., retail refrigerators and coolers, or display refrigerators and coolers. In embodiments, the tracking and / or controlling functions benefit from the use of weighing elements designed to stably support the weight of shelving arrangements, including, without limitation, weighing bars. In exemplary embodiments, the weighing bars are designed so that a pair of weighing bars can stably support a shelving assembly comprising a single shelving bay or a pair of back-to-back shelving bays. In exemplary embodiments, the single-load-cell weighing elements are designed so that four weighing elements can stably support a shelving assembly comprising a single shelving bay or a pair of back-to-back shelving bays. In embodiments, a pair of single-load-cell weighing elements are interchangeable with a weighing bar. As such, embodiments illustrated and / or described herein as employing weighing bars can also be implemented, e.g., in the alternative, using single-load-cell weighing elements without detriment to the embodiments. For the sake of conciseness the following description does not append an alternative description of using single-load-cell weighing elements instead of the illustrated weighing bars, but this should be construed as limiting any embodiment to elongated weighing bars to the exclusion of pairs of single-load-cell weighing elements.

[0043] In some embodiments, a pair of weighing bars can stably support a shelving arrangement comprising a plurality of laterally-aligned shelving assemblies each comprising a single shelving bay or a pair of back-to-back shelving bays. In some embodiments, each of the laterally-aligned shelving assemblies is supported by a respective pair of weighing bars or array comprising two pairs of single-load-cell weighing elements, wherein intermediate weighing bars or pairs of single-load-cell weighing elements can be shared by adjacent shelving assemblies. The weighing elements at each lateral end of the shelving arrangement are not shared, such that a shelving arrangement comprising n shelving assemblies is supported by a total of n+l weighing bars, i.e., or / / + l pairs of single-load-cell weighing elements.

[0044] Weighing bars are structural elements adapted for supporting the various configurations of shelving arrangements, and each comprises one or mechanical members of appropriate shape and strength (e.g., rigidity) for the functionality of supporting the shelving assemblies. The basic weighing function of the load cells, which can comprise, for example, bending beam load cells, are enabled by the presence of a plurality of support legs on each weighing bar. The weighing bars also include load cells, e.g., one or more load cells per weighing bar, that register the weight of the shelving arrangements and of the products disposed there within, i.e., on shelves installed in the shelving arrangements, and transmit streams of weightmeasurement data points to electronic circuitry for further processing.

[0045] The further processing includes detecting an occurrence of a weight event, where a weight event includes the addition of a product to a shelf, the removal of a product from a shelf, and moving a product within a shelving arrangement (or within a specific shelving assembly or shelving bay). The further processing additionally includes identifying a location within the shelving arrangement where the detected weight event has taken place. Identifying the location is accomplished on the basis of the weight-measurement data points streamed from the load cells, and other sensor inputs are generally unnecessary and therefore eschewed. Identifying the location of the weight-event necessarily includes determining the height of the weight-event - for example, determining the specific shelf upon which the weight-event has occurred. Identifying the location of the weight event may also include identifying a specific location upon the identified shelf. The weighing bars / elements and the load cells installed therewithin are configured to track the total weight of multi-shelf shelving arrangements and not just of individual shelves or unitary bins as in the case of some prior art arrangements, and therefore the electronic circuitry provided with the shelving arrangements is made capable of such determining and identifying.

[0046] As further described in this disclosure, methods and systems have been developed for analyzing a detected weight event and identifying the location and product(s) of the weight event. The disclosed methods and systems use probabilistic approaches that broaden their applicability beyond what the deterministic approaches known in the prior art can achieve. A deterministic approach such as can be found in the prior art can be applied, generally speaking, only in a case where products have assigned locations in which only a homogeneous selection of products can be stored or displayed. A deterministic approach is not appropriate for implementations where it is desired to operate without the restriction of homogeneity of products and rigid assignment of products to locations. Systems known in the art have been described as being unable to processes the identification products misplaced, i.e., not placed according to the strict assignment of product to inventory location, and this is a disadvantage that is overcome by the embodiments disclosed herein.

[0047] Many of the disclosed methods and systems base the analysis, and the determination or identification of the location and product of the weight event, not on nominal weight data as is known in the prior art. Instead, the disclosed methods and systems base the analysis on observable patterns in the streams of weightmeasurement data points streamed from the load cells and not solely, or even not all, on the absolute weight values of respective products.

[0048] For example, it is known in the prior art that load cell measurements can be used to determine that a product, e.g., one known to weigh one kg, has been added to a shelving assembly, and to identify the product based on its assigned location. It has also been claimed in the prior art that a weight-event location, such as a height or specific shelf, can be deterministically identified using that information, i.e., the weight of the product, by busing the change in static weight load data from each individual load cell of a plurality of load cells supporting a shelving bay. However, when such an identification, e.g., of weight-event shelf height, is based solely on nominal weight values registered by load cells supporting the shelving assembly because of added or subtracted torque on a mechanical element of the shelving assembly, there are limitations that the instant embodiments overcome. For example, such a prior art approach cannot distinguish between a heavy product close to a backing panel, i.e., on the ‘rear’ of a shelf from an external user’s point of view, from a lighter-weight product far from the backing panel and closer to the front edge of the shelf. Further, such a prior art approach cannot distinguish between a given product closer to the rear of a high shelf from the same product closer to the front of a lower shelf.

[0049] In contradistinction to the limitations of the prior art, the analysis of the stream of weight-measurement data points as embodied in the instant disclosure, along with the fully enables the process of identifying the height of a weight-event, and of the specific shelf involved. This can be accomplished in a number of ways, as described in greater detail hereinbelow, with most approaches involving identification of a pattern in the stream of weight-measurement data points that can indicate a height or a specific shelf with a high degree of accuracy. The identification can include, and not exhaustively, matching an identified pattern to a pattern in a database or to a pattern Teamed’ from training data / or and actual data. A simple example of pattern identification involves assessing amplitude and / or wavelength of an oscillation in the weight-measurement data points. Assigning a height together with a probabilistic distribution of the likely height, thus describing the probability of each shelf, enables identification of an actual shelf with the highest likelihood of being the ‘correct’ answer.

[0050] In further contradistinction to the limitations of the prior art, the analysis of the stream of weight-measurement data points as embodied in the instant disclosure, together with the aforementioned probabilistic approach, fully enable the process of identifying the product or products of the weight-event, i.e., the product(s) added to, removed from, or moved within, the shelving arrangement, e.g., without the need to invoke an alarm condition to request user interference in replacement of the product or manual identification as has been described in the prior art.

[0051] The identification of a location of a weight-event can be practiced today by installing weighing elements, e.g., load cells in various combinations, under, on, or within each of the individual shelves of a shelving bay or refrigerator. This approach has been described in US Patent Application No. 17 / 098565, filed on November 16, 2020, and published as US 2021 / 0131857; US Patent Application No. 17 / 098565, filed on December 6, 2020, and published as US 2021 / 0148750; US Patent Application No. 17 / 134713, filed on December 28, 2020, and published as US 2021 / 0148751; US Patent Application No. 17 / 789195, filed on December 30, 2020, and published as US 2023 / 0021719; and US Patent Application No. 18 / 438527, filed on February 12, 2024, and published as US 2024 / 0257049; the teachings of the foregoing published applications are incorporated herein by reference in their entirety. This approach overcomes the disadvantages discussed hereinabove of prior-art attempts to determine location and product identity of a weight event using weighing elements for an entire shelving bay, but can require the deployment of an order of magnitude more load cells, which for some implementations would be considered too expensive. For example, in an implementation where it is deemed desirable to identify a location on the x-y plane of a specific shelf where a weight-event can occur, each individual shelf is weight-tracked, together with the products displayed thereupon, by four load cells. According to the present embodiments, a typical shelving bay (e.g., a so-called ‘gondola’ bay) with five shelves can be served by two weighing bars, each comprising two load cells or even one load cell, which represents a substantial reduction in the cost of the load cells, and accompanying savings in wiring and installation cost and complexity, relative to prior-art technical solutions. Further, according to embodiments disclosed herein, applying the present disclosure to arrays of shelving bays, i.e., shelving assemblies and shelving arrangements can also be beneficial. In some embodiments, a shelving assembly comprising two back-to-back five-shelf shelving bays can be supported and weight-tracked by a pair of weighing bars each comprising two load cells (or even one load cell), or by four single-load-cell weighing elements, thus yielding an even greater cost reduction relative to the priorart technical solutions. Still further, shelving bays or two-bay shelving assemblies placed side-by-side can ‘share’ weighing bars, as will be discussed in greater detail below with reference to Figs. 11 and 12, such that for every n shelving assemblies, only n+ 1 weighing bars (or pairs of single-load-cell weighing elements) are required. In an exemplary 2 x 10 array of shelving bays, the tracking of changes in weight on the 100 individual shelves can be performed, according to the present embodiments, with 22 load cells or in some implementations only 11 load cells.

[0052] In some of the embodiments, a reduction in the number of load cells can contribute to enabling the deployment of thinner but more expensive planar load cells (e.g., as illustrated in Fig. 4B) instead of the less expensive bending beam load cells (e.g., as illustrated in Fig. 4A) commonly used in industry. Examples of suitable planar load cells can be found in US Patent No. 11092477, the teachings of which are incorporated herein by reference in their entirety. The use of planar load cells, which can have thicknesses under 10 mm or under 5 mm, can advantageously reduce the height added by deploying weighing elements under a shelving bay. In contrast, bending-beam load cells are commonly thicker than 10 mm, or thicker than 20 mm, or thicker than 30mm.

[0053] The inventors have found that a detailed analysis of the streams of weightmeasurement data points generated by the weighing elements or by the load cells installed therein can yield a determination of the height of the weight-event and / or an identification of the shelf on which the weight-event occurred. In embodiments, determination or identification is probabilistic, where any one or more of a number of factors may serve as parameters of the probabilistic determination. In one example, an analysis program can detect a signature characteristic in the stream of weightmeasurement data points in accordance with the identity of the shelf comprising the weight-event location. The signature characteristic can include, for example, amplitude and / or period of an oscillation in the data-point stream generated by the weight-event, optionally including time delays for propagation of oscillations vertically through the shelving bay. Such oscillations occur when using load cells to measure changes in weight, and are primarily due to mechanical vibration in the load cells themselves, in the shelf and / or in the products standing on the shelf, and / or in a product added to a shelf or removed therefrom. An analysis program according to embodiments can be trained to detect patterns in the weight-measurement data points and to compare the pattern to a pattern in a database, e.g., of historical or measured patterns, or to data generated for use in the training.

[0054] In some embodiments, the analysis program uses a probability density function to generate a probabilistic distribution comprising a range of possible heights, with respective probabilities assigned to specific heights. In an example, a centroid of the probabilistic distribution is determined to be the height, i.e., the most likely height, of the weight-event. In another example, a shelf nearest the centroid is identified as the shelf upon which the weight-event occurred, i.e., the most likely candidate to be the shelf on which the weight-event occurred. In still other examples, , other parameters of the probabilistic distribution are used in the determination instead of or in addition to the centroid, such as, for example, a statistical variance. In some examples, it can be that two shelves are identified as equally likely (to be the weightevent shelf), or as both being sufficiently above a threshold probability; in other examples, a single shelf is sufficiently more likely than the other shelves so as to be determinable, i.e., assignable, as being the weight-event shelf.

[0055] In addition to determining a z-axis height of the weight-event, and / or identifying the shelf of the weight-event, it can be desirable to further identify the location by determining where on the shelf the weight-event occurred. This can be accomplished, as is known in the art, by further analyzing the streams of weightmeasurement data points received from the weighing elements or the load cells installed therein, and applying a linear function or a probability density function such as, for example, a bivariate normal distribution to determine, e.g., probabilistically, x- and y-axis coordinates on an x-y plane defined by a surface of the identified shelf, where the x-axis represents a width of the shelf and the y-axis represents the depth from the front of the shelf to the back. In some implementations, it may be deemed sufficient to identify only the x-axis coordinate on the shelf. In such cases, the weighing elements can be configured to include only once load cell each, and the single load cells of each weighing bar of a pair of weighing bars can be adequate when the determining, e.g., probabilistically, is of the x-axis coordinate of the shelf.

[0056] As is known in the art, it is possible to identify the product added, removed or moved based entirely on the location of the weight-event, and especially in retail or inventory locations where care is taken to ensure that products are located where expected, e.g., in accordance with a product plan such as a ‘planogram’. In such implementations, determining the weight-event location is tantamount to identifying the product, because it can involve a simple look-up function.

[0057] In other implementations, a retail establishment or inventory-management facility, e.g., a warehouse, may prefer to not depend entirely on a location-product assignment or ‘lookup’ to identify products removed from shelves, added to shelves, or moved within shelving arrangements. Instead, products involved in weight-events can be identified, inter alia, from the weight-measurement data generated by weighing bars or the load cells installed therein. According to embodiments disclosed herein, the further processing of the weight-measurement data points by electronic circuitry can also include further analyzing the streams of weight-measurement data points to identify the product or products involved in the weight-event.

[0058] The term “electronic circuitry” as used herein means one or more computing devices configured for receiving, processing, analyzing, storing and / or transmitting data. Electronic circuitry or such a computing device should be understood to include any or all of (and not exhaustively): one or more processors, one or more computer- readable media, e.g., transient and / or non-transient storage media, e.g., media containing program instructions for execution by the one or processors as well as data storage, communications arrangements for connections with other networked or nonnetworked computing devices, one or more power sources and / or a connection to a power source, and firmware and / or software. The term “communications arrangements” or similar terms as used herein mean any wired connection or wireless connection via which data communications can take place. Non-limiting and non- exhaustive examples of suitable technologies for providing communications arrangements include any short-range point-to-point communication system such as IrDA, RFID (Radio Frequency Identification), TransferJet, Wireless USB, DSRC (Dedicated Short Range Communications), or Near Field Communication; wireless networks (including sensor networks) such as: ZigBee, EnOcean; Wi-fi, Bluetooth, TransferJet, or Ultra-wideband; and wired communications bus technologies such as . CAN bus (Controller Area Network, Fieldbus, FireWire, HyperTransport and InfiniBand.

[0059] As with the determination of location, product identification can be performed probabilistically, and can be accomplished without input from sensors other than the load cells in the weighing elements. Any one or more of a number of factors can be used, e.g., in building a ranked list of products most likely to be the weight-event product(s), including, and not exhaustively: product location plans (planograms), a database of product weight distribution history, a previous mapping of products on a shelf or shelves stored by and / or accessible to the analysis program, and mechanical characteristics of the product as reflected in oscillations in weight-measurement data points immediately following the weight-event.

[0060] Referring now to the figures, and in particular to Figs 1 A and IB, a shelving arrangement 500 according to embodiments includes a shelving bay 300, which, for example, can be a conventional, commercial shelving bay, e.g., a ‘gondola bay’, comprising a plurality of shelves 90 and a pair of weighing bars 100. In the illustrative example of Figs. 1A and IB, the shelving bay 300 includes five shelves 90, which happens to be a common configuration used in retail stores and is not intended to limit the scope of the embodiments or set an upper or lower limit for the number of shelves in the plurality of shelves. The shelving bay 300 of Figs. 1 A and IB also includes a back panel 80 and a pair of uprights 85. Respective brackets 10 supporting the shelves 90 are hung on or otherwise joined to the uprights 85, said joining being optionally but not necessarily reversible. A shelving-bay base 97 supports the weight of the uprights 85, the back panel 80, the shelves 90, and the brackets 10.

[0061] An upper surface of each of the shelves 90 defines an x-y plane at a z-axis height, e.g., a height above a floor of a facility (or any other height chosen for the ‘origin’ of the z-axis) where the shelving arrangement 500 stands. The basis used for the height need not be the floor as long as the selection of the basis is consistent for purposes of identifying the location of a weight-event. As seen in Fig. 1 A, height zi corresponds to the upper surface of the first shelf 90i, height z corresponds to the upper surface of the second shelf 90z, and so on. As used herein and in the appended claims, the term ‘x- plane’ refers specifically to a portion of a plane corresponding to the surface of a shelf. Thus, an ‘x- plane’ refers to the segment, e.g., rectangular segment, of an x-y plane that is bounded by the edges of the shelf. The ‘x-y plane’ shown in Fig. IB as being at z-axis height zj corresponds to the top shelf 90s, and refers to the area defined by the x-axis width and y-axis depth of the shelf 90s.

[0062] The shelving bay base 97 also supports, as shown in Fig. IB, the aggregate weight of products 70 disposed on all of the shelves 90 of the shelving bay 300. Products 70 may be arranged on shelves 90 in rows or columns, or placed non- homogeneously whether by plan, i.e., for display purposes, or not, e.g., after a period of consumer interaction with the products 70.

[0063] The weight of the shelving bay 300 including all components (e.g., shelves 90, brackets 10, uprights 85, back panel 80 and base 97) and products 70 is supported by the pair of weighing bars 100. Each of the weighing bars 100 comprises one or more load cells and is configured to generate, e.g., by the load cell(s), respective streams of weight-measurement data points corresponding to the weight of the shelving bay 300 and to the products 70 disposed on the shelves 90 of the weighing bay 300. As can be seen in Fig. IB, each weighing bar 100 comprises and is supported by two legs 102. The design of the weighing bar 100 and of the legs 102 is such that the shelving bay 300 is adequately and stably supported by the combined four legs of the two weighing bars 100. It is possible, but not necessary, to add more legs and / or to add a bracing element joining the two weighing bars 100.

[0064] Reference is now made to Figs 1C and ID, a shelving arrangement 500 according to embodiments includes a shelving bay 300, which, for example, can be a conventional, commercial shelving bay, e.g., a ‘gondola bay’, comprising a plurality of shelves 90 and an array of single-load-cell weighing elements 98 (i.e., weighing elements each comprising a single load cell). In the example of Figs. 1C-1D, the array comprises exactly four single-load-cell weighing elements 98. In the illustrative example of Figs. 1C and ID, the shelving bay 300 includes five shelves 90, which happens to be a common configuration used in retail stores and is not intended to limit the scope of the embodiments or set an upper or lower limit for the number of shelves in the plurality of shelves. The shelving bay 300 of Figs. 1C and ID also includes a back panel 80 and a pair of uprights 85. Respective brackets 10 supporting the shelves 90 are hung on or otherwise joined to the uprights 85, said joining being optionally but not necessarily reversible. A shelving-bay base 97 supports the weight of the uprights 85, the back panel 80, the shelves 90, and the brackets 10.

[0065] An upper surface of each of the shelves 90 defines an x-y plane at a z-axis height, e.g., a height above a floor of a facility (or any other height chosen for the ‘origin’ of the z-axis) where the shelving arrangement 500 stands. The basis used for the height need not be the floor as long as the selection of the basis is consistent for purposes of identifying the location of a weight-event. As seen in Fig. 1C, height zi corresponds to the upper surface of the first shelf 90i, height Z2 corresponds to the upper surface of the second shelf 90z, and so on. As used herein and in the appended claims, the term "x-y plane’ refers specifically to a portion of a plane corresponding to the surface of a shelf. Thus, an "x-y plane’ refers to the segment, e.g., rectangular segment, of an x-y plane that is bounded by the edges of the shelf. The "x-y plane’ shown in Fig. ID as being at z-axis height zj corresponds to the top shelf 90s, and refers to the area defined by the x-axis width and y-axis depth of the shelf 90s.

[0066] The shelving bay base 97 also supports, as shown in Fig. ID, the aggregate weight of products 70 disposed on all of the shelves 90 of the shelving bay 300. Products 70 may be arranged on shelves 90 in rows or columns, or placed non- homogeneously whether by plan, i.e., for display purposes, or not, e.g., after a period of consumer interaction with the products 70.

[0067] The weight of the shelving bay 300 including all components (e.g., shelves 90, brackets 10, uprights 85, back panel 80 and base 97) and products 70 is supported by the two pairs of weighing elements 98. Each of the weighing elements 98 comprises one or more load cells 150, and in preferred embodiments comprises exactly one load cell 150, and is configured to generate, e.g., by the load cell(s), respective streams of weight-measurement data points corresponding to the weight of the shelving bay 300 and to the products 70 disposed on the shelves 90 of the weighing bay 300. The design of the weighing elements 98 is such that the shelving bay 300 is adequately and stably supported in combination by the two weighing elements 98.

[0068] We now refer to Figs. 2A-2E, which show various views of exemplary weighing bars 100 according to embodiments. Figs. 2A and 2B are respective side and end views of a non-limiting example of a weighing bar 100 comprising legs 102 characterized by having self-leveling feet, enabled in this example by bearings 104. Fig. 2C is an end view of a weighing bar 100 with adjustable-foot legs 102, and a broader profile than the weighing bar 100 of Figs. 2A and 2B.

[0069] An upwardly open, U-shaped receiving member 110 extends front-to-back along most of the length of the weighing bar 100; in other designs, not shown, the receiving member 110 extends all of the length, or even more than the length, of the weighing bars 100. The receiving member 110 is installed above the load cells 150 and is configured to receive a corresponding load-bearing member of the shelving-bay base 97. The weight of the shelving bay 300 is transferred to the load cells 150 from the receiving member 110 to the one or more load cells 150 via rubber shock absorbers 118.

[0070] Figs. 2C and 2D are schematic side views of exemplary weighing bars 100 with the side plate 108 removed to show an exemplary arrangement of the load cells 150. Fig. 2D shows a weighing bar 100 comprising two load cells 150 and Fig. 2D shows a weighing bar 100 of an alternative design to that of Fig. 2D, comprising a single load cell 150. In both designs, four rubber shock absorbers 118 are provided per load cell 150. The load cells 150 in the non-limiting examples of Figs. 3 and 4 are beam load cells, also known as bending-beam load cells. While this type of load cell lends itself well to the schematically illustrated internal structure of the weighing bars 100, the use of other types of load cells is also within the scope of the invention. The skilled artisan will understand that practicing the disclosure may include changing one or more of the design elements shown in Figs. 2A-2D while upholding the overall design principles.

[0071] In embodiments, a kit (not shown) is provided for supporting and weighing a shelving bay. An exemplary kit comprises a pair of elongated weighing bars 100 adapted for jointly supporting and weighing a shelving bay.

[0072] Reference is now made to Figs. 3 A-3D, which show various examples of single-load-cell weighing elements 98 according to embodiments. Fig. 3A, 3B and 3C show schematic illustrations of exemplary single-load-cell weighing elements 98 each comprising a single load cell 150 installed in a housing 142 adapted for residing on a floor. The housing element 140 typically has a maximum dimension between 5 cm and 20 cm, in accordance with the requirements of the implementation. An attached load-receiving element 140 is positioned to receive a portion of a shelving-bay base 97 or other relevant portion of a shelving bay, such as, for example, a leg (not shown). Transferring the load of a shelving arrangement 500 to the load cell 150 can be accomplished in different ways, depending on the specific design of the receiving element and / or the design of the shelving arrangement 150. In some non-limiting examples, as shown in Figs. 3 A and 3B, an adapter 141 installed in the receiving element 140 has an upper surface higher than an upper surface of the receiving element 140 such a base portion 97 of the shelving arrangement 500 sits atop the adapter 141. In another non-limiting example., a ‘nail-like’ adapter 143 is installed within the receiving element 140 such that a leg of the shelving assembly 500 is received within the receiving element 140 and transfers the load to the nail-like adapter 143. The specific mechanical design of the housing 142, receiving element 140 and adapter 142, 143 are subject to change by the designer who comes to implement the disclosure.

[0073] Fig. 4A shows an exemplary bending-beam load cell 150, illustrated here to provide a non-limiting example of what is meant by the term. Fig. 4B shows an exemplary planar load cell 150, also illustrated here to provide a non-limiting example of what is meant by the term.

[0074] We now refer to Fig. 5, which shows a schematic block diagram of electronic circuitry 40 (or, equivalently, ‘controller’ or ‘computing unit’) associated with, and optionally disposed within, a shelving arrangement 500. In embodiments, any or all of the features and functions of the electronic circuitry 40 as described here can be housed in any one of a computing unit disposed in the shelving arrangement 500; in an ‘external’ computer within the retail location or inventory-management facility in which the shelving arrangement 500 is located; and / or one or more remote computing devices, e.g., in the cloud. For purposes of illustration, the description places all of the functions and features in the electronic circuitry 40 disposed within the shelving arrangement 500.

[0075] The electronic circuitry 40 comprises computing equipment and ancillary equipment configured for receiving streams of weight-measurement data points from weighing bars 100 or the load cells 150 installed therein, for monitoring said streams and detecting weight-events indicating addition to, removal from, or movement of, a product 70 on a shelf 90 in the shelving arrangement 500 (specifically, on one or more shelves 90 in a shelving bay 300), analyzing said streams of data points to determine a location of a weight-event including a height thereof (and / or identifying a shelf upon which the weight-event) and / or to identify the specific product(s) 70 associated with the weight-event, and to interface further with a retail transaction system and / or an inventory management system to transfer therebetween information about products 70 and weight-events. Depending on location customization, the electronic circuitry 40 can include any or all of (and not exhaustively): one or more computer processors 55, computer-readable storage media 58, 59, and a communications module 57. The computer-readable program storage media 58, 59 can include transient and / or transient storage, and can include one or more storage units, all in accordance with desired functionality and design choices. Some or all of the computer-readable program storage media 58, 59 can be cloud-based even if the programmed functions are carried out locally. In embodiments, the program storage 58 can be used for storing program instructions e.g., of the analysis program, in firmware and / or software, for execution by the one or more processors 55; operating and / or historical data and databases supporting the programmed functions related to operation of the shelving arrangements 500 can be stored in the data storage module 59. The communications module 57 is configured to establish communications links with weighing elements 100, 98 or with the load cells 150 installed therein via communications arrangements 91, to communications modules 57 of computing units 40 in other shelving arrangements 500 or other shelving bays 300 via communications arrangements 92, to external computing devices in the retail inventory management location via communications arrangements 93, and / or to cloud-based computers or storage devices via communications arrangements 94. In some embodiments, not all of the illustrated components of the electronic circuitry 40 are provided. In some embodiments, not all of the communications arrangements are provided.

[0076] Referring now to Fig. 6A, a computer-implemented method is disclosed for using a weighing-enabled shelving arrangement 500. The term ‘using’ in this context means using in retail operation and / or in inventory management. The term ‘computer- implemented’ means that the method is carried out by electronic circuitry 40 such as that discussed above with respect to Fig. 5, and / or by another computing unit of the retail or inventory-management facility, and / or by a remote computing unit, e.g., a cloud computing unit in communication with the weighing elements 100, 98 and / or with the electronic circuitry 40. According to the method, the weighing-enabled shelving arrangement 500 may comprise as shown, for example, in Figs. 1 A and IB, a plurality of shelves 90 distributed vertically within a shelving bay 300 and a pair of weighing bars 100 arranged to jointly support the shelving bay 300 and bear the weight thereof. An upper surface of each shelf 90 defines an x-y plane at a respective z-axis height, where the x-y plane is characterized by an x-axis width and a j'-axis depth as illustrated in Fig. IB. Each weighing bar 100 comprises at least one load cell 150. In some embodiments, each weighing bar 100 comprises a plurality of load cells 150. In some embodiments, each weighing bar 100 comprises exactly one load cell 150. In some embodiments, the weighing bars 100 are configured such that the weight of the shelving bay 300 is borne by the respective one or more load cells 150 installed in each of the weighing bars 100. According to the method, the weighing-enabled shelving arrangement 500 may comprise as shown, for example, in Figs. 1C and ID, a plurality of shelves 90 distributed vertically within a shelving bay 300 and an array of single-load-cell weighing elements 98 arranged to jointly support the shelving bay 300 and bear the weight thereof.

[0077] As illustrated by the flow chart in Fig. 6A, the method comprises at least the two steps SOI, S02:

[0078] Step SOI includes receiving respective streams of weight-measurement data points corresponding to the shelving bay 300 and to products disposed on shelves 90 therewithin.

[0079] Step S02 includes analyzing the received weight-measurement data points to determine a z-axis height of a location of a weight-event in which a product is added to, removed from, or moved within the shelving bay 300. In some embodiments, the determining of the z-axis height includes identifying a shelf, i.e., the shelf 90 on which a weight-event occurred. In some embodiments, a program stored in and executed by the electronic circuitry 40 is trained to associate patterns in weightmeasurement data points with respective z-axis heights of shelves 90. In some embodiments, the determining includes determining x- and z- coordinates of the location of the weight event. In some embodiments, the determining includes determining x-, y- and z- coordinates of the location of the weight-event. In some embodiments, the determining includes identifying a location on the x-y plane of the identified shelf 90. In some embodiments, the determining is probabilistic. In some embodiments, the determining includes selecting the z-axis height from a range of z- axis heights corresponding to a portion of a probabilistic distribution comprising the location. In some embodiments, the determining is carried out without input from sensors that are not weight sensors. In some embodiments, the only weight measurements used in the determining are from the load cells 150 installed in the weighing bars 100 or single-load-cell weighing elements 98. In some embodiments, the analyzing of Step S02 includes assessing at least one of an amplitude and a wavelength of an oscillation in the received weightmeasurement data points caused by the weight-event. In some embodiments, the analyzing includes detecting a pattern in the received weight-measurement data points and comparing the pattern to a pattern in a database. In some embodiments, the analyzing includes detecting a pattern in the received weight-measurement data points and comparing the pattern to data used for training the analysis program.

[0080] Figs. 7 and 8 present graphs of changes in weight vs. time, illustrating exemplary aspects of the analyzing of Step S02.

[0081] In a first example, the graph of Fig. 7 shows curves representing the weightmeasurement data points received from four respective load cells that are arranged in four weighing elements of a single-bay shelving arrangement. In the graph, -axis values are ‘normalized’ to show the total weight before the weight-event as ‘zero’. The added solid-line curve is an aggregation of the data of the four load cells LCO- LC3. The respect weight-measurement data points of the four lad cells are shown individually by the dashed-line curves.

[0082] From time = zero, until time ~ 1.75, one sees typical minor fluctuation in weight values of less than 50 gm due to environmental noise such as vibrations. The weight event begins with a rapid increase in weight, as is common with product removals; the rapid increase reflects the force of a customer placing a hand on the product about to be removed. Here the total -weight curve peaks at about 350 gm about half a second later, and then the weight rapidly decreases before reaching .

[0083] In the example of Fig. 7, the respective shapes of the individual load cell curves in the post-removal phase show oscillations recognizable by the analysis program, e.g., by using pattern-recognition, as being most likely caused by a weightevent that occurred on the top (uppermost) shelf of five shelves installed in a shelving bay. The weight-event is determined to be a removal of a product weighing approximately 500 gm. Continuing the analysis of the data points, the analysis program is used, in embodiments, to identify a location on the x-y plane of the top shelf based again on the weight-measurement data points received from the four load cells LC0-LC3. It can be seen, for example, that the greatest increase in weight during the initial rapid increase, and the greatest decrease in weight during the post-removal phase, are those shown by LCO and LC3, in that order, and LC1 and LC2 are the least effected by the weight event. The analysis program applies, for example, a linear function or a probability density function such as, for example, a bivariate normal distribution to determine, e.g., probabilistically, x- and j'-axis coordinates on the x-y plane.

[0084] In a second example, Fig. 8 shows curves representing the weightmeasurement data points received from four respective load cells that are arranged in four weighing elements of a single-bay shelving arrangement. The solid-line curves are generated by a product removal weight-event on the bottom shelf of the shelving bay (labeled as shelf5, using a shelf-numbering convention opposite to that used in the other figures herein), and the dashed-line curves are generated by a product removal weight-event on the top shelf of the shelving bay (labeled here as shelf 1). It can be understood from the graph that weight-events occurring on different shelves generate respective patterns of oscillations that are different from each other, e.g., in amplitude and frequency. These differences enable the various techniques used by the analysis program to determine, probabilistically or otherwise, a z-axis height location and specific shelf location of the weight-event.

[0085] Referring again to the method illustrated in Fig. 6A: in some embodiments, the method additionally comprises Step S03, as illustrated in Fig. 6B.

[0086] Step S03 includes further analyzing the received weight-measurement data points to identify the product(s) 70 corresponding to the weight-event. In some embodiments, the identifying of the product 70 is carried out without input from sensors that are not the weighing elements 100, 98 or the load cells 150 installed therein. In some embodiments, the identified product is one of a group of non- homogeneous products 70 disposed on the shelves in the shelving bay. In some embodiments, the identifying of the product 70 is not based solely on the determined coordinates of the location of the weight-event. In some embodiments, the identifying of the product is probabilistic.

[0087] In some embodiments, the further analyzing of Step S03 includes accessing information in at least one of a product database, a planogram, and a mapping of products performed before the weight-event as part of the identification process performed by the analysis program, e.g., in order to build a ranked list of products most likely to be the weight-event product(s),.

[0088] In some embodiments, the weighing-enabled shelving arrangement 500 used for implementing the method additionally comprises the back panel 80 and the pair of uprights 85 shown in Fig. IB, such that each of the plurality of shelves 90 is supported by a pair of brackets 10 respectively joined to the pair of uprights 85.

[0089] In some embodiments, the shelving arrangement 500 used for implementing the method further comprises one or more additional shelving bays 300 and either of: one or more additional weighing bars 100 each comprising at least one load cell 150, or an additional pair of single-load-cell weighing elements 98. Each additional shelving bay 300 includes a plurality of shelves 90 distributed vertically therewithin, and (as disclosed above with respect to the first shelving bay 300) an upper surface of each shelf 90 defines an x-y plane at a respective z-axis height. In such embodiments, the electronic circuitry 40 is further programmed to carry out the method with respect to the one or more additional shelving bays 300 and the one or more additional weighing bars 100 or pairs of single-load-cell weighing elements 98, i.e., one computing unit can handle implementing the method for the entire multi-bay shelving arrangement 500.

[0090] Referring now to Fig. 9, a method is disclosed for using a weighing-enabled shelving arrangement 500, e.g., in retail operation and / or in inventory management. The method is carried out by electronic circuitry 40 such as that discussed above with respect to Fig. 5, and / or by another computing unit of the retail or inventorymanagement facility, and / or by a remote computing unit, e.g., a cloud computing unit in communication with the weighing elements 100, 98 and / or with the electronic circuitry 40. The weighing-enabled shelving arrangement 500 comprises, according to the method and as shown, for example, in Figs. 1 A and IB, a plurality of shelves 90 distributed vertically within a shelving bay 300 and a pair of weighing bars 100 ( or four single-load-cell weighing elements 98) arranged to jointly support the shelving bay 300 and bear the weight thereof. An upper surface of each shelf 90 defines an x-y plane at a respective z-axis height, where the x-y plane is characterized by an x-axis width and a -axis depth as illustrated in Fig. IB.

[0091] As illustrated by the flow chart in Fig. 9, the method comprises at least the three steps Sil, S12 and S13.

[0092] Step S13 includes receiving respective streams of weight-measurement data points corresponding to the shelving bay 300 and to products 70 disposed on shelves 90 therewithin.

[0093] Step S14 includes detecting an occurrence of a weight-event in which a product 70 is added to or removed from the shelving bay 300, or moved within the shelving bay 300 or shelving arrangement 500. The detection is in response to a change over time in values of the received weight-measurement data points.

[0094] Step S15 includes analyzing the received weight-measurement data points to identify the product 70 added to or removed from the shelving bay 300, or moved within the shelving bay 300 or shelving arrangement 500. In some embodiments, the analyzing includes determining a range of z-axis heights comprising a location of the weight-event. In some such embodiments, the range of z-axis heights corresponds to a portion of a probabilistic distribution comprising the location. In some embodiments, where the plurality of shelves 90 comprises at least three shelves 90, the range of z- axis heights includes the z-axis heights of exactly two shelves 90. In some embodiments, the range of z-axis heights includes the z-axis height of exactly one shelf 90. In some embodiments, the analyzing includes determining a z-axis height of a location of the weight-event, and that z-axis height corresponds to a single shelf 90. In some embodiments, the analyzing additionally includes determining a location on the respective x-y plane defined by the upper surface of the single shelf 90. In some embodiments, the analyzing additionally includes determining an x- coordinate of the location of the weight-event. In some embodiments, the analyzing additionally includes determining x- and y- coordinates of the location of the weight-event.

[0095] In some embodiments, the analyzing includes assessing at least one of an amplitude and a wavelength of an oscillation in the received weight-measurement data points caused by the weight-event. In some embodiments, the analyzing includes detecting a pattern in the received weight-measurement data points and comparing the pattern to training data.

[0096] The discussion above of the graphs of Fig. 7 and 8 is applicable, mutatis mutandis, to the performance of Step S15.

[0097] In some embodiments, the analyzing does not include determining a z-axis height of the location of the weight-event.

[0098] In some embodiments, the identifying is carried out without input from sensors that are not weight sensors. In some embodiments, the only weight measurements used in the identifying are from the weighing elements or the load cells installed therein. In some embodiments, the identifying of the product 70 is not based solely on the determined coordinates of the location of the weight-event.

[0099] In some embodiments, at least one of the identifying of the product 70 and the determining of the range of z-axis heights is probabilistic. We now refer to Figs. 10A, IB, 11 and 12, in which all of the exemplary shelving arrangements are shown as including elongated weighing bars 100; as discussed hereinabove, the shelving arrangements can alternatively be provided using single-load-cell weighing elements 98, as shown, e.g., in Figs. 1C and ID, using the general equivalence of one pair of single-load-cell elements 98 replacing each individual weighing bar 100.

[0100] In embodiments, a shelving arrangement 500 includes an array of n shelving assemblies 400, e.g., one shelving assembly, or two shelving assemblies, or at least 5 or at least 10 shelving assemblies. A shelving assembly 400 can either comprise a single shelving bay 300 as shown, e.g., in Figs. 1 A and IB, or can comprise two shelving bays 300 arranged back-to-back, e.g., as shown in Fig. 10A. Thus, the single shelving bay 300 of Figs. 1A and IB is an example of an array of n shelving assemblies 400 where n=l and the lone shelving assembly 400 comprises only one shelving bay 300.

[0101] In the illustrative example of Fig. 10 A, both the ‘A’ -side shelving bay 300A and the ‘B’-side shelving bay 300B happen to have the same number of shelves 90. (The A-side and B-side are indicated by the arrow in Fig. 10A.) Although this is a common configuration in many supermarkets, the embodiments are not limited by this example and the two back-to-back bays can have any number of bays, including different numbers of bays on each side. The ‘A’ -side shelves 90A are arranged in Fig. 10A with upper surfaces at respective z-axis heights ZA, and the ‘B’-side shelves 90B are arranged at respective z-axis heights ZB, While the z-axis heights shown in Fig. 10A happen to be the same for both shelving bays 300A, 300B, practicing the embodiment is not limited to observing this convention, and the two sets of shelves 90A, 90B can be set at any suitable heights ZA, ZB.

[0102] In some implementations, back-to-back shelving bays 300 of a single shelving assembly 400 share some components, such as, for example, the back panel 80 and double-sided uprights 85. In addition, as shown in Fig. 10A, the back-to-back shelving bays 300 of the shelving assembly 400 can share a single pair of weighing bars 100 configured to support, in combination, the weight of both shelving bays 300A, 300B. The weighing bars 100 are configured to transmit respective streams of weight-measurement data points corresponding to the array of shelving assemblies 400 and to products 70 (products not shown in Figs. 10A-12) disposed on shelves 90 therewithin. Fig. 10B shows the weighing bar 100 of Fig 10A with the side plate 108 removed to show an exemplary arrangement of the two load cells 150. Each of the weighing bars 100 is arranged so that a first load cell 100 installed in an A-side end of the weighing bar 100 is at least partly disposed under the A-side shelving bay 300A, and the other load cell 150 is installed in a B-side end of the weighing bar 100 is disposed under the B-side shelving bay 300B.

[0103] In some implementations, when n is greater than 1, the shelving assemblies 400 are arranged side-by-side and share common components such as uprights 85 and weighing bars 100. Thus, according to embodiments, a shelving arrangement 500 comprising an array of n shelving assemblies 400 arranged side-by-side can include as few as n+ weighing bars. Each of the shelving bays 300 stands on, or is supported by, two different weighing bars 100, and each weighing bar 100 except the first and last partly supports two different shelving bays 300.

[0104] Fig. 11 shows an exemplary shelving arrangement 500 including an array of three shelving assemblies 400, arranged side-by-side, each of the shelving assemblies 400 comprising a single shelving bay 300. The shelving arrangement 500 includes four weighing bars 100 arranged as described above in a ‘shared’ configuration. For example, it can be seen that the second weighing bar IOO2 partly supports the combined weights of the first shelving bay 300i and of the second shelving bay 3002. All of the shelving bays 300 are shown with the same configuration of shelves 90 installed at the same heights; while this is a configuration commonly used in this kind of shelving, the illustration is not intended to be limiting, and different shelving bays 300 within the same shelving arrangement 500, even when arranged side-by-side, can have different quantities of shelves 90 distributed therein and at different z-axis heights.

[0105] Fig. 12 shows a second exemplary shelving arrangement 500 including an array of three shelving assemblies 400, arranged side-by-side, each of the shelving assemblies 400 comprising two back-to-back shelving bays 300, e.g., based on the two-bay shelving assembly 400 of Fig. 10A. Again, the shelving arrangement 500 includes four weighing bars 100 arranged in the ‘shared’ configuration. Each of the weighing bars 100 is arranged similarly to that shown in Fig. 10B except that the weighing bar 100 is partly supporting two adjacent shelving assemblies 400 comprising four shelving bays 300. Thus, for example, the second weighing bar IOO2 partly supports the combined weights of the following shelving bays: 300IA, 3002A, 300IB, and 300ZB (Shelving bay 300ZB is not visible in the view shown in Fig. 12, as most of the B-side bays are obscured). In some embodiments, each weighing bar 100 comprises a plurality of load cells 150, and in some embodiments, each weighing bar 100 comprises exactly one load cell.

[0106] In the example of Fig. 12, each weighing bar 100 comprises exactly two load cells 150. A first load cell 150 is installed under the A-side end of the weighing bar IOO2 and is at least partly disposed under shelving bays 300IA and 300IA. A second load cell 150 is installed under the A-side end of the weighing bar IOO2 and is at least partly disposed under shelving bays 300IA and 3002A.

[0107] The shelving arrangements 500 of Figs. 10A, 11 and 12 all include respective computing units 40 (‘electronic circuitry’), which may not be shown in the figures. The respective computing units 40 are programmed to detect a change over time in values of weight-measurement data points indicating a weight-event. The computing units 40 are further programmed to analyze the values to determine a location of the weight-event, including the identification of a specific shelving bay 300, and a specific shelf or pair of shelves in that specific bay 300.

[0108] In some embodiments, determining of the location of the weight event by the computing units 40 includes determining x- and z- coordinates of the location, or determining x-, y- and z- coordinates of the location. In some embodiments, the determining includes identifying a location on the x-y plane of said identified shelf. The order of the analysis is not critical, and in some embodiments, the analysis program determines the shelf or z-axis height before identifying the specific shelving bay 300 or more specific x- and / or y- coordinates on the shelf 90.

[0109] In some embodiments, the determining by the computing units 40 is probabilistic. The determining may include selecting the z-axis height from a range of z-axis heights corresponding to a portion of a probabilistic distribution comprising the location. In some embodiments, the determining is carried out without input from sensors that are not weight sensors, and in some embodiments, the only weight measurements used in the determining are from the load cells.

[0110] In some embodiments, the analyzing by the computing units 40 includes assessing at least one of an amplitude and a wavelength of an oscillation in the weight-measurement data points following the weight-event. The analyzing can include detecting a pattern in the weight-measurement data points and comparing the

[0111] T1 pattern to a pattern in a stored database and or to data used to train the analysis program.

[0112] In some embodiments, the computing unit 40 is additionally programmed to further analyze the weight-measurement data points to identify the product 70 corresponding to the weight-event. In some embodiments, the identifying of the product 70 is probabilistic and is not necessarily based solely on the coordinates determined as being the location of the weight-event - the further analyzing can include accessing at least one of a product database, a planogram, and a mapping of products performed before the weight-event. In some embodiments, the product 70 is identified without input from sensors that are not the weighing bars 100 or the load cells 150. In some embodiments, the product 70 is identified from amongst a group of non-homogeneous products 70 disposed on the shelves 90 in the specific shelving bay 300 or across different shelving bays 300.

[0113] We now refer to Figs. 13, 14A and 14B, any or all of the teachings hereinabove are applicable, mutatis mutandis, to refrigerators in which products are stored and / or displayed on shelves. The refrigerators are shown as including elongated weighing bars 100; as with the open-bay shelving arrangements of the preceding paragraphs, the refrigerators can alternatively be provided using single-load-cell weighing elements 98, as shown, e.g., in Figs. 1C and ID, using the general equivalence of one pair of single-load-cell elements 98 replacing each individual weighing bar 100.

[0114] Fig. 13 shows an exemplary retail-enabled refrigerator 200 according to embodiments. A vertical array of five shelves 90i .. 90s are used to display products 70 for direct sale, each shelf 90 having an upper surface at a respective z-axis height zi .. Z5. The upper surface of each shelf 90 defines an x-y plane (only the x-axis is indicated in Fig. 13, as the -axis is obscured).

[0115] The refrigerator 200 additionally includes a plurality of weighing bars 100 arranged bear the weight of the refrigerator, where ‘bearing the weight’ means bearing the weight except that of the weighing bars themselves. The weighing bars 100 each comprise one or more load cells 150, and the load cells 150, or another component of the weighing bars 100 configured to transmit respective streams of weight-measurement data points corresponding to the refrigerator 200 (excluding the weighing bars themselves) and to products 70 disposed on shelves 90. The refrigerator also includes a computing unit 40 (not shown) programmed to receive the respective streams of weight-measurement data points transmitted from the weighing bars 100, to detect an occurrence of a weight-event in which a product 70 is added to, removed from, or moved within the refrigerator 200, and analyze said the weight-measurement data points to identify the product 70.

[0116] In embodiments, the identifying is carried out without input from sensors that are not weight sensors, and the only weight measurements used in identifying the product 70 are those transmitted from the weighing bars 100 or the load cells 150.

[0117] The exemplary retail-enabled refrigerator 200 of Fig. 13 further comprises a refrigeration unit 250, a retail transaction apparatus 230 and an optional door 220.

[0118] Referring now to Figs. 14A and 14B, a refrigerator 200 may comprise more than one shelving section 350, where the shelving sections 350 are distributed horizontally in the refrigerator 200. Each shelving section 350 comprises a vertical array of shelves 90 on which products 70 are stored and / or displayed. The multisection refrigerator 200 may have a single base 210 supporting the weight of the refrigerator 200 and in turn supported by the weighing bars 100. In some implementations, each shelving section 350 can have a separate base 210. Both Figs. 14A and 14B shown examples of multi-section refrigerators 200 in which all of the sections 350 have the same number of shelves 90 at the same z-axis heights. In other implementations, not illustrated, the number of shelves 90 and / or the respective z-axis heights of the shelves can vary between shelving-sections 350 with no effect on the ability of the computing unit 40 to identify the product 70 involved in the weightevent.

[0119] The number of weighing bars 100 used with a multi-section refrigerator 200 can vary, where options include: using a single pair of weighing bars 100, supporting shelving section 350 by a pair of adjacent weighing bars 100 as shown in the example of Fig. 14A, and an intermediate configuration in which at least one weighing bar 100 is used in an intermediate position as shown in the example of Fig. 14B.

[0120] When identifying the product 70 involved in a weight-event within a multisection refrigerator 200 such as one of the exemplary refrigerators shown in Figs. 14A and 14B, the identifying performed by the computer unit 40 can include determining the shelving section 350 in which the weight-event occurred.

[0121] For any of the refrigerators 200 shown in Figs. 13, 14A and 14B, identifying the product 70 can includes determining upon which shelf 90 the weight-event occurred. The process of identifying the product 70 includes, in some embodiments, analyzing the weight-measurement data points to determine a range of z-axis heights comprising the location of the weight-event. The range of z-axis heights can correspond to a portion of a probabilistic distribution comprising the location. In embodiments, at least one of the identifying of the product and the determining of the range of z-axis heights is probabilistic, as opposed to deterministic. In some embodiments, both the identifying of the product and the determining of the range of z-axis heights are probabilistic.

[0122] In some embodiments, identifying the product 70 includes determining an x- coordinate of the location of the weight-event and, in some embodiments, includes determining both x- and y- coordinates.

[0123] In some embodiments, analyzing the weight-measurement data points includes assessing the amplitude and / or wavelength of oscillations detected in the weightmeasurement data points following the weight-event. The analyzing can include detecting a pattern in the weight-measurement data points and comparing the pattern to data used to train the analysis program run by the computing unit 40, and or to a pattern in a database.

[0124] In some embodiments, identifying of the product is not based solely on the determined coordinates of the location of the weight-event.

[0125] Any of the methods and / or method steps disclosed herein, whether disclosed as methods and method steps or as program instructions executed by a computing unit 40 of a shelving arrangement 500 or refrigerator 200, can be combined in any suitable way, and any such combination is within the scope of the disclosed embodiments.

[0126] Any of the methods and / or method steps disclosed herein can be carried out by (or in response to an instruction by) a computing unit 40 of a shelving arrangement 500 or refrigerator 200.

[0127] The present disclosure includes, and not exhaustively, the following inventive concepts, numbered 1-117 for convenient reference.

[0128] Inventive concept 1. A computer-implemented method of using a weighing- enabled shelving arrangement, the shelving arrangement comprising (i) a plurality of shelves distributed vertically within a shelving bay, an upper surface of each shelf defining an x-y plane at a respective z-axis height, the x-y plane being characterized by an x-axis width and a j'-axis depth, and (ii) an array of weighing elements arranged to jointly support the shelving bay and bear the weight thereof, each weighing element comprising at least one load cell, the method comprising: a. receiving respective streams of weight-measurement data points corresponding to the shelving bay and to products disposed on shelves therewithin; and b. analyzing the received weightmeasurement data points to determine a z-axis height of a location of a weight-event in which a product is added to, removed from, or moved within the shelving bay.

[0129] Inventive concept 2. The method of inventive concept 1, wherein the analyzing includes detecting a pattern in the weight-measurement data points.

[0130] Inventive concept 3. The method of inventive concept 2, wherein the analyzing includes detecting a pattern in the weight-measurement data points and comparing the detected pattern to a pattern in a database.

[0131] Inventive concept 4. The method of either one of inventive concepts 2 or 3, wherein the analyzing includes using training data to determine a z-axis height.

[0132] Inventive concept 5. The method of any one of the preceding inventive concepts, wherein the analyzing includes detecting a pattern in the weightmeasurement data points and comparing the detected pattern to training data.

[0133] Inventive concept 6. The method of any one of inventive concepts 3 to 5, wherein the comparing does not include comparing, in isolation, a specific weight value.

[0134] Inventive concept 7. The method of any one of the preceding inventive concepts, wherein the analyzing includes identifying indicators, in the weightmeasurement data points, of a mechanical response of a shelf to the weight event.

[0135] Inventive concept 8. The method of any one of the preceding inventive concepts, wherein the analyzing includes identifying indicators of a mechanical response to the weight event of the product added to, removed from, or moved within the shelving bay. Inventive concept 9. The method of any one of the preceding inventive concepts, wherein determining the z-axis height is not based on a weight of the product added to, removed from, or moved within the shelving bay.

[0136] Inventive concept 10. The method of any one of the preceding inventive concepts, wherein determining the z-axis height does not take into account a measured weight of the product added to, removed from, or moved within the shelving bay.

[0137] Inventive concept 11. The method of any one of the preceding inventive concepts, wherein determining the z-axis height does not take into account an expected, stored, or known weight of the product added to, removed from, or moved within the shelving bay.

[0138] Inventive concept 12. The method of any one of the preceding inventive concepts, wherein determining the z-axis height does not take into account an identity of the product added to, removed from, or moved within the shelving bay.

[0139] Inventive concept 13. The method of any one of the preceding inventive concepts, wherein determining the z-axis height is probabilistic.

[0140] Inventive concept 14. The method of any one of the preceding inventive concepts, wherein determining the z-axis height includes selecting the z-axis height from a range of z-axis heights corresponding to a portion of a probabilistic distribution comprising the location.

[0141] Inventive concept 15. The method of any one of the preceding inventive concepts, wherein determining the z-axis height includes identifying a shelf.

[0142] Inventive concept 16. The method of any one of the preceding inventive concepts, wherein determining the z-axis height includes identifying a shelf, and identifying the shelf includes selecting the shelf from a plurality of shelves identified as having respective probabilities for the weight-event being located thereupon.

[0143] Inventive concept 17. The method of any one of the preceding inventive concepts, wherein the determining is carried out without input from sensors that are not weight sensors.

[0144] Inventive concept 18. The method of any one of the preceding inventive concepts, wherein the only weight measurements used in the determining are from the load cells.

[0145] Inventive concept 19. The method of any one of the preceding inventive concepts, wherein the analyzing includes assessing at least one of an amplitude and a wavelength of an oscillation in said weight-measurement data points caused by the weight-event.

[0146] Inventive concept 20. The method of any one of the preceding inventive concepts, wherein the determining includes determining x- and z- coordinates of the location of the weight event.

[0147] Inventive concept 21. The method of any one of the preceding inventive concepts, wherein the determining includes determining x-, y- and z- coordinates of the location of the weight-event. Inventive concept 22. The method of any one of inventive concepts 2 to 4, wherein the determining includes identifying a location on the x-y plane of said identified shelf.

[0148] Inventive concept 23. The method of any of the preceding inventive concepts, wherein each weighing element comprises an elongated weighing bar comprising a respective plurality of load cells.

[0149] Inventive concept 24. The method of any one of inventive concepts 1 to 22, wherein each weighing element comprises exactly one load cell.

[0150] Inventive concept 25. The method of any one of the preceding inventive concepts, additionally comprising: further analyzing said weight-measurement data points to identify the product corresponding to the weight-event.

[0151] Inventive concept 26. The method of inventive concept 25, wherein the further analyzing accesses at least one of a product database, a planogram, and a mapping of products performed before the weight-event.

[0152] Inventive concept 27. The method of either one of inventive concepts 25 or 26, wherein the identifying of the product is carried out without input from sensors that are not the weighing bars or the load cells installed therein.

[0153] Inventive concept 28. The method of any one of inventive concepts 25 to 27, wherein the identified product is one of a group of non-homogeneous products disposed on the shelves in the shelving bay.

[0154] Inventive concept 29. The method of any one of inventive concepts 25 to 28, wherein the identifying of the product is not based solely on the determined coordinates of the location of the weight-event.

[0155] Inventive concept 30. The method of any one of inventive concepts 25 to 29, wherein the identifying of the product is probabilistic.

[0156] Inventive concept 31. A shelving arrangement, comprising: a. a shelving bay including a plurality of shelves distributed vertically therewithin, an upper surface of each shelf defining an x-y plane at a respective z-axis height; b. an array of weighing elements each comprising at least one load cell, the weighing elements arranged to support the shelving bay and bear the weight thereof, and configured to transmit respective streams of weight-measurement data points corresponding to the shelving bay and to products disposed on shelves therewithin; and c. a computing unit programmed to carry out the method of any one of inventive concepts 1 to 30.

[0157] Inventive concept 32. The shelving arrangement of inventive concept 31, additionally comprising a back panel and a pair of uprights associated therewith, wherein each of the plurality of shelves is supported by a pair of brackets respectively joined to the pair of uprights.

[0158] Inventive concept 33. The shelving arrangement of either one of inventive concepts 31 or 32, wherein the weighing elements are configured such that the weight of the shelving bay is borne by the load cells installed in the weighing elements.

[0159] Inventive concept 34. The shelving arrangement of any one of inventive concepts 31 to 33, wherein the computing unit is trained to associate patterns in weight-measurement data points with respective z-axis heights of shelves.

[0160] Inventive concept 35. The shelving arrangement of any one of inventive concepts 31 to 34, further comprising (i) one or more additional shelving bays, each additional shelving bay including a plurality of shelves distributed vertically therewithin, an upper surface of each shelf defining an x-y plane at a respective z-axis height, and (ii) one or more additional sets of weighing elements, each additional set of weighing elements comprising a single elongated weighing bar comprising a plurality of load cells or a plurality of single-load-cell weighing elements each comprising a single load cell, wherein the computing unit is further programmed to carry out the method with respect to the one or more additional shelving bays and the one or more additional sets of weighing elements.

[0161] Inventive concept 36. A shelving arrangement, comprising: a. a shelving bay including (i) a back panel and a pair of uprights associated therewith, (ii) a plurality of shelves distributed vertically within the shelving bay, each shelf supported by a pair of brackets respectively joined to the pair of uprights, an upper surface of the shelf defining a plane at a respective height, and (iii) a shelving-unit base supporting the weight of the uprights, back panel, shelves and brackets; and b. an array of weighing elements arranged to support the shelving bay and bear the weight thereof, and configured to generate respective streams of weight-measurement data points corresponding to the shelving bay and to products disposed on shelves therewithin.

[0162] Inventive concept 37. The shelving arrangement of inventive concept 36, wherein the array of weighing elements comprises a pair of elongated weighing bars each comprising one or more load cells, Inventive concept 38. The shelving arrangement of inventive concept 36, wherein each of the weighing bars comprises, for each of the load cells therein, four rubber shock absorbers mediating between the shelving bay and the respective load cells.

[0163] Inventive concept 39. The shelving arrangement of either one of inventive concepts 37 or 38, wherein each of the weighing bars comprises exactly two legs, and the one or more load cells are disposed between the two legs.

[0164] Inventive concept 40. The shelving arrangement of inventive concept 39, wherein each of the legs includes a self-leveling foot.

[0165] Inventive concept 41. The shelving arrangement of any one of inventive concepts 37 to 40, wherein the shelving bay additionally comprises a shelving-bay base, and each of the weighing bars comprises an upwardly open, U-shaped upper member configured to receive a portion of the shelving-bay base.

[0166] Inventive concept 42. The shelving arrangement of any one of inventive concepts 37 to 41, wherein the pair of weighing bars are not physically connected to each other.

[0167] Inventive concept 43. A weighing bar according to any one of inventive concepts 38 to 41.

[0168] Inventive concept 44. The shelving arrangement of inventive concept 36, wherein the array of weighing elements comprises at least four single-load-cell weighing elements each comprising a single load cell.

[0169] Inventive concept 45. A shelving arrangement, comprising: a. an array of n shelving assemblies, n being greater than or equal to one, each shelving assembly comprising at least one shelving bay and not more than two shelving bays, each shelving bay including a respective plurality of shelves distributed vertically therewithin, an upper surface of each shelf defining a x-y plane at a respective z-axis height, the x-y plane being characterized by an x-axis width and ay-axis depth; b. an array of n+1 sets of weighing elements, each set of weighing elements comprising a single elongated weighing bar comprising a plurality of load cells or a plurality of single-load-cell weighing elements each comprising a single load cell, the weighing elements arranged to support the shelving assemblies and bear the weight thereof, and configured to transmit respective streams of weight-measurement data points corresponding to the array of shelving assemblies and to products disposed on shelves therewithin, the arranging being such that each shelving assembly rests on two consecutive sets of weighing elements; and c. a computing unit programmed to detect a change over time in values of said weight-measurement data points indicating a weight-event, and to analyze said values to determine a location of the weight-event, wherein said determining includes identifying a shelving bay and a shelf therewithin.

[0170] Inventive concept 46. The shelving arrangement of inventive concept 45, wherein n is greater than one.

[0171] Inventive concept 47. The shelving arrangement of either one of inventive concepts 45 or 46, wherein each of the shelving assemblies comprises two back-to- back shelving bays, and each of the sets of weighing elements is arranged so that a first load cell installed in a first portion thereof is at least partly disposed under a first one of the two shelving bays and a second load cell installed in a second portion thereof is disposed under a second one of the shelving bays.

[0172] Inventive concept 48. The shelving arrangement of either one of inventive concepts 45 or 46, wherein each of the shelving assemblies comprises exactly one shelving bay, and each of the sets of weighing elements is arranged so that each of the plurality of load cells installed therein is at least partly disposed under the one respective shelving bay.

[0173] Inventive concept 49. The shelving arrangement of any one of inventive concepts 45 to 48, wherein the analyzing includes determining a z-axis height of the weight-event.

[0174] Inventive concept 50. The shelving arrangement of any one of inventive concepts 45 to 49, wherein the analyzing includes determining a z-axis height of the weight-event, and the determining includes assessing at least one of an amplitude and a wavelength of an oscillation in said weight-measurement data points caused by the weight-event.

[0175] Inventive concept 51. The shelving arrangement of any one of inventive concepts 45 to 49, wherein the analyzing includes determining a z-axis height of the weight-event by detecting a pattern in the weight-measurement data points and comparing the detected pattern to a pattern in a database.

[0176] Inventive concept 52. The shelving arrangement of any one of inventive concepts 45 to 49, wherein the analyzing includes determining a z-axis height of the weight-event using training data.

[0177] Inventive concept 53. The shelving arrangement of any one of inventive concepts 45 to 49, wherein the analyzing includes determining a z-axis height of the weight-event by identifying indicators, in the weight-measurement data points, of a mechanical response of a shelf to the weight event.

[0178] Inventive concept 54. The shelving arrangement of any one of inventive concepts 45 to 49, wherein the analyzing includes determining a z-axis height of the weight-event by identifying indicators of a mechanical response to the weight event of the product added to, removed from, or moved within the shelving bay. Inventive concept 55. The shelving arrangement of any one of inventive concepts 45 to 49, wherein the analyzing includes determining a z-axis height of the weight-event, and the determining is not based on a weight of the product added to, removed from, or moved within the shelving bay.

[0179] Inventive concept 56. The shelving arrangement of any one of inventive concepts 45 to 49, wherein the analyzing includes determining a z-axis height of the weight-event, and the determining is not based on a a measured weight of the product added to, removed from, or moved within the shelving bay.

[0180] Inventive concept 57. The shelving arrangement of any one of inventive concepts 45 to 49, wherein the analyzing includes determining a z-axis height of the weight-event, and the determining is not based on an expected, stored, or known weight of the product added to, removed from, or moved within the shelving bay.

[0181] Inventive concept 58. The shelving arrangement of any one of inventive concepts 45 to 49, wherein the analyzing includes determining a z-axis height of the weight-event, and the determining is not based on an identity of the product added to, removed from, or moved within the shelving bay.

[0182] Inventive concept 59. The shelving arrangement of any one of inventive concepts 45 to 58, wherein determining the location of the weight-event is carried out without input from sensors that are not weight sensors.

[0183] Inventive concept 60. The shelving arrangement of any one of inventive concepts 45 to 59, wherein the only weight measurements used in the determining of the location of the weight-event are from the load cells.

[0184] Inventive concept 61. The shelving arrangement of any one of inventive concepts 45 to 60, additionally comprising: further analyzing the weight-measurement data points to identify the product corresponding to the weight-event.

[0185] Inventive concept 62. The shelving arrangement of any one of inventive concepts 45 to 61, wherein the determining of the location of the weight-event is probabilistic. Inventive concept 63. The shelving arrangement of any one of inventive concepts 45 to 63, wherein each of the weighing elements comprises, for each of the load cells therein, four rubber shock absorbers mediating between the shelving bay and the respective load cells.

[0186] Inventive concept 64. The shelving arrangement of any one of inventive concepts 45 to 63, wherein each of the sets of weighing elements comprises a weighing bar comprising exactly two legs, and the plurality of load cells are disposed between the two legs.

[0187] Inventive concept 65. The shelving arrangement of any one of inventive concepts 45 to 64, wherein each of the weighing elements includes a self-leveling foot.

[0188] Inventive concept 66. The shelving arrangement of any one of inventive concepts 45 to 64, wherein each of the weighing elements includes a manual leveling mechanism.

[0189] Inventive concept 67. A computer-implemented method of using a weighing- enabled shelving arrangement, the shelving arrangement comprising (i) a plurality of shelves distributed vertically within a shelving bay, an upper surface of each shelf defining an x-y plane at a respective z-axis height, the x-y plane being characterized by an x-axis width and a j'-axis depth, and (ii) an array of weighing elements each comprising one or more load cells, the weighing elements arranged to jointly support the shelving bay and bear the weight thereof, the method comprising: a. receiving respective streams of weight-measurement data points corresponding to the shelving bay and to products disposed on shelves therewithin; b. responsively to a change over time in values of said weight-measurement data points, detecting an occurrence of a weight-event in which a product is added to, removed from, or moved within the shelving bay; and c. analyzing said weight-measurement data points to identify the product.

[0190] Inventive concept 68. The method of inventive concept 67, wherein the analyzing includes determining a range of z-axis heights comprising a location of the weight-event.

[0191] Inventive concept 69. The method of inventive concept 68, wherein the range of z-axis heights corresponds to a portion of a probabilistic distribution comprising the location. Inventive concept 70. The method of either one of inventive concepts 68 or 69, wherein the plurality of shelves comprises at least 3 shelves, and the range of z- axis heights includes the z-axis heights of exactly two shelves of the plurality of shelves.

[0192] Inventive concept 71. The method of any one of inventive concepts 67 to 70, wherein the analyzing includes detecting a pattern in the weight-measurement data points.

[0193] Inventive concept 72. The method of any one of inventive concepts 67 to 70, wherein the analyzing includes detecting a pattern in the weight-measurement data points and comparing the detected pattern to a pattern in a database.

[0194] Inventive concept 73. The method any one of inventive concepts 68 to 70, wherein the analyzing includes using training data to determine a z-axis height.

[0195] Inventive concept 74. The method of any one of inventive concepts 68 to 73, wherein the determining of the range of z-axis heights comprising a location of the weight-event analyzing includes detecting a pattern in the weight-measurement data points and comparing the detected pattern to training data.

[0196] Inventive concept 75. The method of inventive concept 74, wherein the comparing does not include comparing, in isolation, a specific weight value.

[0197] Inventive concept 76. The method of any one of inventive concepts 71 to 75, wherein the analyzing includes identifying indicators, in the weight-measurement data points, of a mechanical response of a shelf to the weight event.

[0198] Inventive concept 77. The method of any one of inventive concepts 71 to 76, wherein the analyzing includes identifying indicators of a mechanical response to the weight event of the product added to, removed from, or moved within the shelving bay. Inventive concept 78. The method of any one of inventive concepts 68 to 77, wherein determining the z-axis height is not based on a weight of the product added to, removed from, or moved within the shelving bay.

[0199] Inventive concept 79. The method of any one of inventive concepts 68 to 78, wherein determining the z-axis height does not take into account a measured weight of the product added to, removed from, or moved within the shelving bay.

[0200] Inventive concept 80. The method of any one of inventive concepts 68 to 79, wherein determining the z-axis height does not take into account an expected, stored, or known weight of the product added to, removed from, or moved within the shelving bay.

[0201] Inventive concept 81. The method of any one of inventive concepts 68 to 80, wherein determining the z-axis height does not take into account an identity of the product added to, removed from, or moved within the shelving bay.

[0202] Inventive concept 82. The method of any one of inventive concepts 67 to 81, wherein the analyzing does not include determining a z-axis height of the location of the weight-event.

[0203] Inventive concept 83. The method of any one of inventive concepts 67 to 82, wherein the identifying is carried out without input from sensors that are not weight sensors.

[0204] Inventive concept 84. The method of any one of inventive concepts 67 to 83, wherein the only weight measurements used in the identifying are from the weighing elements or the load cells installed therein.

[0205] Inventive concept 85. The method of any one of inventive concepts 67 to 84, wherein the analyzing includes assessing at least one of an amplitude and a wavelength of an oscillation in said weight-measurement data points caused by the weight-event.

[0206] Inventive concept 86. The method of any one of inventive concepts 67 to 85, wherein the analyzing includes detecting a pattern in said weight-measurement data points and comparing the pattern to training data.

[0207] Inventive concept 87. The method of any one of inventive concepts 68 to 86, wherein the identifying of the product is not based solely on the determined coordinates of the location of the weight-event.

[0208] Inventive concept 88. The method of any one of inventive concepts 68 to 87, wherein at least one of the identifying of the product and the determining of the range of z-axis heights is probabilistic.

[0209] Inventive concept 89. A shelving arrangement, comprising: a. a shelving bay including a plurality of shelves distributed vertically therewithin, an upper surface of each shelf defining an x-y plane at a respective z-axis height; b. an array of weighing elements each comprising one or more load cells, the weighing elements arranged to jointly support the shelving bay and bear the weight thereof, and configured to transmit respective streams of weight-measurement data points corresponding to the shelving bay and to products disposed on shelves therewithin; and c. a computing unit programmed to carry out the method of any one of inventive concepts 67 to 88.

[0210] Inventive concept 90. The shelving arrangement of inventive concept 89, additionally comprising a back panel and a pair of uprights associated therewith, wherein each of the plurality of shelves is supported by a pair of brackets respectively joined to the pair of uprights.

[0211] Inventive concept 91. The shelving arrangement of either one of inventive concepts 89 or 90, wherein the weighing elements are configured such that the weight of the shelving bay is borne by the load cells installed in the weighing elements.

[0212] Inventive concept 92. The shelving arrangement of any one of inventive concepts 89 to 91, wherein the computing unit is trained to associate patterns in weight-measurement data points with respective z-axis heights of shelves.

[0213] Inventive concept 93. The shelving arrangement of any one of inventive concepts 89 to 92, further comprising (i) one or more additional shelving bays, each additional shelving bay including a plurality of shelves distributed vertically therewithin, an upper surface of each shelf defining an x-y plane at a respective z-axis height, and (ii) one or more additional sets of weighing elements, each additional set of weighing elements comprising a single elongated weighing bar comprising a plurality of load cells or a plurality of single-load-cell weighing elements each comprising a single load cell, wherein the computing unit is further programmed to carry out the method with respect to the one or more additional shelving bays and the one or more additional sets of weighing elements bars.

[0214] Inventive concept 94. A weighing-enabled refrigerator, comprising: a. a refrigerator including one or more shelving sections distributed horizontally therein, each shelving section comprising a vertical array of shelves, an upper surface of each shelf defining an x-y plane at a respective z-axis height; b. an array of weighing elements each comprising at least one load cell, the weighing elements arranged to support the refrigerator and bear the weight thereof, and configured to transmit respective streams of weight-measurement data points corresponding to the refrigerator and to products disposed on shelves therewithin; and c. a computing unit programmed to i. receive said respective streams of weight-measurement data points, ii. responsively to a change over time in values of said weight-measurement data points, detect an occurrence of a weight-event in which a product is added to, removed from, or moved within the refrigerator, and iii. analyze said weightmeasurement data points to identify the product.

[0215] Inventive concept 95. The refrigerator of inventive concept 94, comprising exactly one shelving section.

[0216] Inventive concept 96. The refrigerator of inventive concept 94, comprising a plurality of shelving sections, wherein the identifying includes determining the shelving section in which the weight-event occurred.

[0217] Inventive concept 97. The refrigerator of any one of inventive concepts 94 to 96, wherein the identifying includes determining a shelf upon which the weight-event occurred.

[0218] Inventive concept 98. The refrigerator of any one of inventive concepts 94 to 96, wherein the identifying includes analyzing said weight-measurement data points to determine a range of z-axis heights comprising a location of the weight-event.

[0219] Inventive concept 99. The refrigerator of inventive concept 98, wherein the range of z-axis heights corresponds to a portion of a probabilistic distribution comprising the location.

[0220] Inventive concept 100. The refrigerator of any one of inventive concepts 94 to

[0221] 99, wherein the identifying includes determining an x- coordinate of the location of the weight-event.

[0222] Inventive concept 101. The refrigerator of any one of inventive concepts 94 to

[0223] 100, wherein the identifying includes determining x- and j'- coordinates of the location of the weight-event.

[0224] Inventive concept 102. The refrigerator of any one of inventive concepts 94 to

[0225] 101, wherein the identifying is carried out without input from sensors that are not weight sensors.

[0226] Inventive concept 103. The refrigerator of any one of inventive concepts 94 to

[0227] 102, wherein the only weight measurements used in the identifying are from the weighing elements or the load cells installed therein.

[0228] Inventive concept 104. The refrigerator of any one of inventive concepts 94 to

[0229] 103, wherein the analyzing includes assessing at least one of an amplitude and a wavelength of an oscillation in said weight-measurement data points caused by the weight-event. Inventive concept 105. The refrigerator of any one of inventive concepts 94 to

[0230] 104, wherein the analyzing includes detecting a pattern in said weight-measurement data points and comparing the pattern to training data.

[0231] Inventive concept 106. The refrigerator of any one of inventive concepts 94 to

[0232] 105, wherein the identifying of the product is not based solely on the determined coordinates of the location of the weight-event.

[0233] Inventive concept 107. The refrigerator of any one of inventive concepts 98 to

[0234] 106, wherein at least one of the identifying of the product and the determining of the range of z-axis heights is probabilistic.

[0235] Inventive concept 108. The refrigerator of any one of inventive concepts 94 to

[0236] 107, additionally comprising a retail transaction apparatus.

[0237] Inventive concept 109. A kit comprising a pair of elongated weighing bars adapted for jointly supporting and weighing a shelving bay, wherein each of the weighing bars comprises: a. A plurality of load cells, b. a spaced-apart plurality of leg supports, c. an upper frame member adapted to receive a portion of a load-bearing member of a base of the shelving-bay, and d. for each of the one or more load cells, a lower frame member comprising a first portion supported by a corresponding leg support and second portion having the respective load cell fixed thereto, and a plurality of rubber shock absorbers mediating between the upper frame member and the respective lower frame member.

[0238] Inventive concept 110. The kit of inventive concept 109, wherein the load cells comprise bending-beam load cells.

[0239] Inventive concept 111. The kit of inventive concept 109, wherein the load cells comprise planar load cells.

[0240] Inventive concept 112. The kit of either one of inventive concepts 109 or 110, wherein the load cells are fixed to undersides of the respective lower frame members.

[0241] Inventive concept 113. The kit of any one of inventive concepts 109 to 111, wherein the upper frame member comprises an upwardly open, U-shaped member.

[0242] Inventive concept 114. The kit of any one of inventive concepts 109 to 112, wherein each of the leg supports comprises a manual leveling mechanism.

[0243] Inventive concept 115. The kit of any one of inventive concepts 109 to 113, wherein each of the leg supports comprises a self-leveling foot.

[0244] Inventive concept 116. Any one of inventive concepts 1 to 115, wherein the load cell(s) comprise bending-beam load cells. Inventive concept 117. Any one of inventive concepts 1 to 115, wherein the load cell(s) comprise planar load cell(s).

[0245] The present invention has been described using detailed descriptions of embodiments thereof that are provided by way of example and are not intended to limit the scope of the invention. The described embodiments comprise different features, not all of which are required in all embodiments of the invention. Some embodiments of the present invention utilize only some of the features or possible combinations of the features. Variations of embodiments of the present invention that are described and embodiments of the present invention comprising different combinations of features noted in the described embodiments will occur to persons skilled in the art to which the invention pertains.

Claims

CLAIMS1. A computer-implemented method of using a weighing-enabled shelving arrangement, the shelving arrangement comprising (i) a plurality of shelves distributed vertically within a shelving bay, an upper surface of each shelf defining an x-y plane at a respective z-axis height, the x-y plane being characterized by an x-axis width and a j'-axis depth, and (ii) an array of weighing elements arranged to jointly support the shelving bay and bear the weight thereof, each weighing element comprising at least one load cell, the method comprising: a. receiving respective streams of weight-measurement data points corresponding to the shelving bay and to products disposed on shelves therewithin; and b. analyzing the received weight-measurement data points to determine a z- axis height of a location of a weight-event in which a product is added to, removed from, or moved within the shelving bay, wherein the analyzing includes detecting a pattern in the weight-measurement data points.

2. The method of claim 1, wherein the analyzing additionally includes comparing the detected pattern to training data.

3. The method of either one of claims 1 or 2, wherein the analyzing additionally includes comparing the detected pattern to a pattern in a database.

4. The method of either one of claims 2 or 3, wherein the comparing does not include comparing, in isolation, a specific weight value.

5. The method of any one of the preceding claims, wherein the analyzing includes identifying indicators, in the weight-measurement data points, of a mechanical response of a shelf to the weight event.

6. The method of any one of the preceding claims, wherein the analyzing includes identifying indicators of a mechanical response to the weight event of the product added to, removed from, or moved within the shelving bay.

7. The method of any one of the preceding claims, wherein the analyzing includes assessing at least one of an amplitude and a wavelength of an oscillation in said weight-measurement data points caused by the weight-event.

8. The method of any one of the preceding claims, wherein determining the z-axis height is not based on a weight of the product added to, removed from, or moved within the shelving bay.

9. The method of any one of the preceding claims, wherein determining the z-axis height does not take into account an identity of the product added to, removed from, or moved within the shelving bay.

10. The method of any one of the preceding claims, wherein determining the z-axis height includes selecting the z-axis height from a range of z-axis heights corresponding to a portion of a probabilistic distribution comprising the location.

11. The method of any one of the preceding claims, additionally comprising: further analyzing said weight-measurement data points to identify the product corresponding to the weight-event.

12. The method of claim 11, wherein the identifying of the product is carried out without input from sensors that are not the weighing bars or the load cells installed therein.

13. A shelving arrangement, comprising: a. a shelving bay including a plurality of shelves distributed vertically therewithin, an upper surface of each shelf defining an x-y plane at a respective z-axis height; b. an array of weighing elements each comprising at least one load cell, the weighing elements arranged to support the shelving bay and bear the weight thereof, and configured to transmit respective streams of weightmeasurement data points corresponding to the shelving bay and to products disposed on shelves therewithin; and c. a computing unit programmed to carry out the method of any one of claims 1 to 12.

14. The shelving arrangement of claim 13, wherein the computing unit is trained to associate patterns in streams of weight-measurement data points with respective z- axis heights of shelves.

15. A shelving arrangement, comprising: a. an array of n shelving assemblies, n being greater than one, each shelving assembly comprising at least one shelving bay and not more than two shelving bays, each shelving bay including a respective plurality of shelves distributed vertically therewithin, an upper surface of each shelf defining a x-y plane at a respective z-axis height, the x-y plane being characterized by an x-axis width and a j'-axis depth; b. an array of n+1 sets of weighing elements, each set of weighing elements comprising a single elongated weighing bar comprising a plurality of load cells or a plurality of single-load-cell weighing elements each comprising a single load cell, the weighing elements arranged to support the shelving assemblies and bear the weight thereof, and configured to transmit respective streams of weight-measurement data points corresponding to the array of shelving assemblies and to products disposed on shelves therewithin, the arranging being such that each shelving assembly rests on two consecutive sets of weighing elements; and c. a computing unit programmed to detect a change over time in values of said weight-measurement data points indicating a weight-event, and to analyze said values to determine a location of the weight-event, wherein said determining includes identifying a shelving bay and a shelf therewithin.

16. The shelving arrangement of claim 15, wherein each of the shelving assemblies comprises two back-to-back shelving bays, and each of the sets of weighing elements is arranged so that a first load cell installed in a first portion thereof is at least partly disposed under a first one of the two shelving bays and a second load cell installed in a second portion thereof is disposed under a second one of the shelving bays.

17. The shelving arrangement of claim 15, wherein each of the shelving assemblies comprises exactly one shelving bay, and each of the sets of weighing elements isarranged so that each of the plurality of load cells installed therein is at least partly disposed under the one respective shelving bay.

18. The shelving arrangement of any one of claims 15 to 17, wherein the analyzing includes determining a z-axis height of the weight-event, and the determining includes assessing at least one of an amplitude and a wavelength of an oscillation in said weight-measurement data points caused by the weight-event.

19. The shelving arrangement of any one of claims 15 to 17, wherein the analyzing includes determining a z-axis height of the weight-event by detecting a pattern in the weight-measurement data points and comparing the detected pattern to a pattern in a database.

20. The shelving arrangement of any one of claims 15 to 17, wherein the analyzing includes determining a z-axis height of the weight-event using training data.

21. The shelving arrangement of any one of claims 15 to 20, wherein the analyzing includes determining a z-axis height of the weight-event by identifying indicators, in the weight-measurement data points, of a mechanical response of a shelf to the weight event.

22. The shelving arrangement of any one of claims 15 to 21, wherein the analyzing includes determining a z-axis height of the weight-event, and the determining is not based on a weight of the product added to, removed from, or moved within the shelving bay.

23. A computer-implemented method of using a weighing-enabled shelving arrangement, the shelving arrangement comprising (i) a plurality of shelves distributed vertically within a shelving bay, an upper surface of each shelf defining an x-y plane at a respective z-axis height, the x-y plane being characterized by an x-axis width and a j'-axis depth, and (ii) an array of weighing elements each comprising one or more load cells, the weighing elements arranged to jointly support the shelving bay and bear the weight thereof, the method comprising: a. receiving respective streams of weight-measurement data points corresponding to the shelving bay and to products disposed on shelves therewithin;b. responsively to a change over time in values of said weight-measurement data points, detecting an occurrence of a weight-event in which a product is added to, removed from, or moved within the shelving bay; and c. analyzing said streams of weight-measurement data points to identify the product, wherein the analyzing includes detecting a pattern in the weight-measurement data points.

24. A shelving arrangement, comprising: a. a shelving bay including a plurality of shelves distributed vertically therewithin, an upper surface of each shelf defining an x-y plane at a respective z-axis height; b. an array of weighing elements each comprising one or more load cells, the weighing elements arranged to jointly support the shelving bay and bear the weight thereof, and configured to transmit respective streams of weight-measurement data points corresponding to the shelving bay and to products disposed on shelves therewithin; and c. a computing unit programmed to carry out the method of claim 23.

25. The shelving arrangement of claim 24, wherein the computing unit is trained to associate patterns in weight-measurement data points with respective z-axis heights of shelves.

26. The shelving arrangement of any one of claims 89 to 92, further comprising (i) one or more additional shelving bays, each additional shelving bay including a plurality of shelves distributed vertically therewithin, an upper surface of each shelf defining an x-y plane at a respective z-axis height, and (ii) one or more additional sets of weighing elements, each additional set of weighing elements comprising a single elongated weighing bar comprising a plurality of load cells or a plurality of single-load-cell weighing elements each comprising a single load cell, wherein the computing unit is further programmed to carry out the method with respect to the one or more additional shelving bays and the one or more additional sets of weighing elements bars.

Images