Patent Grant
11232498
The disclosed embodiments are directed to labeling and distributing products having multiple versions while providing correlation of user ratings with a received product version on a per user basis.
Conventional approaches to the production of consumer products are designed around efficiency and economies of scale, so whether one is manufacturing toothpaste or running a winery producing chardonnay, the objective is to put together a large production run so that everything is cheaper. In the case of labeled bottles of wine, efficiency is achieved by buying or producing a large quantity of bottles of chardonnay, for example, and a large quantity of labels and running a labeling line continuously over an extended period of time. When the labeling of the chardonnay is completed, then one might proceed with labeling a different type of wine, such as cabernet sauvignon. Thus, in conventional approaches to product labeling and distribution, an entire production run of a product is labeled with the same label for consistency and pricing efficiency, because high-volume runs are cheaper.
In conventional approaches to product brand/label design, a company might hire an agency, perhaps providing a creative brief. The agency would come up with brand/labeling ideas, and a decision maker would pick an idea and have the agency develop it. Also, research would be done to avoid competing brands/labeling. In the case of wine, a large winery might do consumer testing, e.g., having 10 people in a room with a 2-way mirror and watching them react to wines on shelf. Such approaches may be more inclined to indulge the predilections of the decision maker, e.g., a winery owner, rather than resulting in an objective and systematic consideration of consumer views.
The disclosed embodiments provide a platform for controlling branding, labeling, and distribution of products, e.g., consumable items, based on machine learning processes derived from user ratings of products randomly labeled with different versions of branding/labeling. The consumable item may be wine, e.g., bottles of wine distributed via a website store or wine club. Panels of defined brand/label characteristics of the wines may be obtained in a systematic manner. The brand/label data may be used to build a user level model which may be used, for example, to establish a user preference profile and provide recommendations to the entity responsible for branding and labeling. The brand/label data may also be used to build Gaussian Mixture Models (GMM) which are a form of machine learning. A GMM model may be used to look at distributions of subsets of data and generate novel data that fits well within a given distribution. Such models can be used to create “algorithmic” branding/labeling, e.g., wine labels. Deep learning models may be used to build neural networks to perform functions such as generating wine labels and performing high-dimensional clustering of wine labeling and branding attributes. The branding/labeling data may be input into dynamic time warping (DTW) alignment calculations which perform mathematical operations to determine how different the characteristic vectors are from one another. The results may be used to build a model of spatial ranking (SR), which is stored. DTW can be used to align vectors which are non-linear and widely distributed and collapse them into the same vector space. This includes calculating an energy cost to collapsing the vectors into the same space, i.e., a measure of how different the vectors are from one another.
In one aspect, the disclosed embodiments provide a method (and corresponding system and software) for labeling and distributing products having multiple versions with recipient version correlation on a per user basis. The method includes storing, on at least one server having a processor and memory, one or more values characterizing a first label, each of the values being associated with a label characteristic of a defined set of label characteristics. The method further includes storing, on the at least one server, one or more values characterizing a second label, each of the values being associated with a label characteristic of the defined set of label characteristics. The method further includes labeling a first group of a product with the first label and a second group of the product with the second label. The method further includes distributing the product from the first group and from the second group to users in response to user orders for the product, wherein each user has an account stored at the at least one server. The method further includes receiving, from the users, user ratings for the product, each of the user ratings being identified as associated with the first group or the second group depending on whether, in the distributing, a user providing each of the user ratings received the product from the first group or from the second group. The method further includes correlating the user ratings with the values of the label characteristics.
Aspects of the disclosed embodiments may include one or more of the following features. The correlating of the user ratings with the values of the label characteristics may include a regression. The correlating of the user ratings with the values of the label characteristics may include a machine learning algorithm. The label characteristics may include one or more of the following: modern vs. traditional, abstract vs. figurative, award medals, color, font size. The first label and the second label may be based on results of a survey sent to a group of consumers. The group of consumers may be selected according to demographic characteristics. The first label and the second label may be produced by varying at least one label characteristic of the defined set of label characteristics of a label selected based on the results of the survey.
In the case of wine labels, each of the values is associated with a label characteristic of a defined set of label characteristics, e.g., font style, font size, font color, category of images (e.g., animals, grape vines, winery chateau, etc.), background color, etc. A number of other characteristics may also be used, for example print techniques (e.g., identifying a label shape, a material, an embossing or debossing technique, etc.).
As shown in
The system 10 further includes a web server 30, having an associated database 35, for hosting a website for distributing the product from the first group and from the second group to customers 40. As shown, a plurality of customers 40 are in communication with the server 30 via a network 45. The customers 40 may operate computing devices (such as, for example, mobile phones, computers, or the like) to interact with the web server 30 in order to place orders for products. In some embodiment, each customer 40 has an associated account with an entity operating the web server 30 allowing the customer's order (and order history) to be tracked and associated with the customer 40. Pursuant to some embodiments, customers may interact with the web server 30 to select or otherwise place orders for products.
In some cases, the website hosting server 30 may be the same server 15 as the one described above for managing label characteristics. The website may host a wine club (and/or a wine sales site), in which case each customer 40 has an account stored at the website server 30. The distribution may involve sending wine bottles labeled with the first or second label randomly to the wine club customers 40. The customers may not be aware that there is more than one version of the label. The system 10 maintains information identifying which customers 40 received which label version.
The system 10 provides for receiving, from the wine club customers 40, user ratings for the product. For example, the customers may be prompted to rate the wine (not the label itself). The customers may be prompted to enter their ratings and any comments in a web form associated with the web server 30 or in response to a comment or feedback form. The user ratings are identified as associated with the first group or the second group depending on whether a customer 40 received the product from the first group or from the second group, respectively. Thus, all of the customers 40 may receive the same product but with one or the other of the two labels being tested (in the disclosed embodiment there are two labels being tested but comparison of three or more labels is also possible, as noted above). In such a case, the distribution of the two different labels among the customers may be done, for example, in a random manner.
The user ratings are correlated with the values of the label characteristics. For example, in a simple case, the user ratings associated with the first and second labels may be compared to determine which label garnered a better reaction from the customers (it is assumed that difference in the user ratings of the wine are due to the customer's reaction to the label, as the product itself is the same). Other types of analysis may also be performed based on the user ratings. For example, the two labels being tested have different label characteristics and the ratings may be correlated with the particular values of the label characteristics, e.g., by performing a regression or by applying a machine learning algorithm. In this way, the effects of each label characteristic, e.g., font size, can be evaluated separately.
Disclosed embodiments provide a systematic method of producing better product labels/brands and, provide techniques by which product labels are systematically made better (e.g., more appealing to the consumer). In some cases, user response (e.g., user ratings and/or comments), may be gathered once a product is in market. A set of brand characteristics and brand subject matter may be defined in which a numeric value is assigned to each characteristic. For example, a product label may be characterized in terms of, e.g., modern/abstract v. traditional/restrained, etc. A code may be assigned to each of these characteristics on produced labels to compare the “DNA” of the label to actual consumer feedback on a per user basis. In other words, the user ratings provided by each individual user can be correlated with a particular product version having a particular set of characteristics.
Pursuant to some embodiments, a plurality of product labels may be generated and sent to consumers as part of a survey, e.g., via a text message to the users' mobile devices. This may be done using software which provides templates for creating surveys and which allows for the specification of demographics, e.g., age, number of children, income, etc., for the survey recipients. Such a survey may be done very soon after generation of the labels, e.g., in less than an hour, and the text message may be sent to many, e.g., thousands, of consumers. The survey may present, e.g., two or three different labels and may indicate to the consumer that they are in a particular demographic from which opinions are being sought. The survey may ask which label is most appealing, which one would the consumer pay $20 for, and/or which label puts the consumer “in a good mood,” etc. The survey may result in quick feedback, e.g., within an hour or so. Several versions, e.g., two or three versions, may be generated of a label which receives high ratings in the survey.
As explained above, products, e.g., wine, may be labeled with label variations A, B, and C generated based on the survey results and may be sold/distributed to consumers. The consumers may not be aware that the product is being distributed with varied labels. For example, a first consumer may receive a product, e.g., a bottle of wine, which has a frog on the label (“Label A”). A second consumer may receive the same bottle wine, i.e., the exact same product, with a different image, such as a newt (“Label B”) or a vine (“Label C), instead of the frog. In such a case, a difference in consumer ratings of the identical wine bottled under the three different label variations may be considered to indicate a preference for the labeling itself. In disclosed embodiments, the wine may be distributed via a direct-to-consumer wine club which is in contact with the end consumer all the time. In such a case, it can be determined who rated a wine with a particular label variation higher. In other words, it can be determined whether there was a perception of quality or taste difference based only on label, e.g., a label with a newt versus one with a frog. This information may be provided to a brand team so that they know what to use on the wine label, i.e., a newt or a frog. Testing conducted in this manner could be used to refine a label even further, e.g., by testing whether there is a difference in user ratings between a label version having a blue newt versus a label version having a red newt.
As explained above, the survey steps may be used in conjunction with a specialized production line on which the same wine is bottled with multiple variations in labeling/packaging. For example, a survey might be used to determine whether displaying an award medal on the label of a wine bottle improves the perception of wine. This could be followed by a production run in which a portion of the bottles have a label with a medal on it, while the remainder have a label without a medal. User ratings may be obtained from users who have received the wine labeled randomly with one of the two label versions. In experiments using the present invention, a 6-7% increase in customer satisfaction for the same wine, same name, same brand, etc., with the only difference being a gold medal on the label. Thus, it has been found that including an award medal on the label provides a significant increase in customer satisfaction but does not increase costs. By way of comparison, the difference in user ratings for a poor-quality merlot and a good merlot might be about 10%.
In disclosed embodiments, an analysis may be made of product branding/labeling using axes of, e.g., new vs. traditional, whimsical vs. serious, etc. The analysis may be used to identify “gaps” in product branding/labeling and distribution. For example, the analysis may find that a winery does not have a California wine that is “whimsical and youthful.”
In disclosed embodiments, production lines may be labeling bottles of a particular wine with different labels and distributing the wine through the wine club managed and accessed via the website. In such a case, data relating to the distribution of the wine and the resulting user ratings may be constantly obtained and analyzed. A customer signing up for a wine club may specify particular preferences, e.g. red wine only, U.S. wine only, which reduces the set of products which is sent to the customer. An algorithm may be used to pick wines, e.g., based on chemical profiles, without regard to brand. Unbeknownst to the customer, the algorithm may select a wine for which there are two different labels, e.g., a wine having a production run of 30,000 bottles, of which 15,000 bottles have “Label A” and the remainder have “Label B,” in a random distribution.
Pursuant to some embodiments, customers 40 are encouraged to rate the wines they receive, (e.g., to allow each customer's preferences to influence which wines the customer receives in the wine club), and also to provide text reviews. Pursuant to some embodiments, the text reviews or comments are analyzed using natural language processing (“NLP”) techniques to extract and classify comments from customers. In some embodiments, trained personnel may be used to review and normalize comments (either aided by or in lieu of automated text analysis). In general, embodiments classify or normalize comments allowing the feedback to be classified into categories associated with the product, such as the print technique (the label shape, etc.), the style (e.g., modern, masculine, bold, dark, light, etc.), the mood (e.g., serious, whimsical, etc.), and the subject (e.g., geographical, historical, cultural, etc.).
In some embodiments, the text reviews are read and interpreted by either (or both) of a trained NLP classifier or by trained personnel, because customers may not use the same terms a wine taster would use (e.g., a comment stating “this wine burned my throat” may be interpreted to mean that the wine has too much alcohol or tannins). Based on the received user ratings and interpreted comments, a series of dashboards may be provided on the wine making side and business side. The business users can pull up the two versions of wine labels and look at the differences, including total user rating score and common words used to describe the two different versions. For example, a wine distributed with a label version without an award medal may be described as “rich” and “fruity,” whereas the same wine distributed with a label which displays an award medal may be described as “high quality and good value.” The business user can see the difference in the user rating scores, e.g., the wine without an award medal on the label may be rated positively by 80% of users, whereas the same wine with an award medal on the label may be rated positively by 86% of users.
In disclosed embodiments, an Internet-based “club” format may be used to distribute products, such as, for example, a wine club. In this context, the distributor has definitive information on the source of user ratings, which can be used to personalize the wines that are distributed to each user based on the user's expressed preferences. In addition, user rating data may be used to evaluate wines and their branding/labeling. By contrast, if a winery distributes wine through a grocery store, then they do not know who the end customer is and whether they will buy wine again. For example, the customer might pay cash or buy the same wine later in another store. In such cases, there is no systematic record of who the customer is.
In disclosed embodiments, products, e.g., wines, having different label versions may each be distributed with a different unique identifier, e.g., a different stock keeping unit (SKU). For example, a SKU may begin with “RWCABCA” to indicate that the product is a red wine, a cabernet, and from California. This may be followed by a code indicating a brand name and then a sequential identifier corresponding to the various test versions, e.g., test labels. In such a case, the different versions of a product, e.g., a wine having different label versions, may be treated as separate products in the warehouse and in the packing line, but to the customer, they may appear to be the same product.
In disclosed embodiments, characteristics of a product other than labeling or branding may be varied and randomly distributed. For example, a wine may be produced in versions which have two different levels of sugar to assess user preference for sweetness. In such a case, a particular level of sugar, e.g., 5 g/l, which is dry but on the sweeter side, may be found to be optimal, whereas European wines typically have 0-3 g/l of sugar. Such tests are helpful in determining what a customer really wants in a wine. For example, a customer may say that they want a “sweet” wine, but in the wine industry that term corresponds to wines with 20 g/l of sugar. It may turn out that the customer actually prefers wines having levels of sugar that are considered dry (not even off-dry) in the wine industry. Thus, this indirect evaluation of wine sweetness may overcome a discrepancy in terminology between the customer and the wine makers and distributors.
As a further example, a wine may be distributed in two versions, one with a cork and one with a screw cap to ascertain whether customers' perceptions of a screw cap have a negative impact on a wine's ratings. As a further example, a wine may be distributed in a version with heavy glass (e.g., 650 g) and a lighter glass (e.g., 500 g) to determine whether it is worth paying the additional cost to use heavy glass.
The randomized distribution of different versions of the same product, while treating the versions as a single product vis-à-vis the customer, allows product characteristics to be evaluated indirectly, which helps to reduce subjective bias. For example, a customer may respond to a direct question regarding their preference for wine bottles with lighter glass or heavier glass by indicating that they do not have a preference or, perhaps, that they prefer lighter glass. However, data from randomized distribution of versions of the same wine bottled with lighter and heavier glass may reveal higher user ratings for heavy glass. Such experiments can easily be run across fifty thousand customers, whereas a typical winery which does not have a “club” customer base would find it difficult to find such a large test population.
Further details of features of some embodiments will now be described by referring to several flow diagrams that describe processes that may be performed by an application performed by a computing device or a group of computing devices, such as a user device, a web server, a host platform, a cloud computing environment, and the like to implement features of the present invention. Reference is now made to
In disclosed embodiments, the standardized data is stored as a model object which is stored in cloud storage, e.g., Amazon S3 Cloud storage. This data is used in processes which are configured to pull down the defined brand/label characteristic panels for particular wines. In disclosed embodiments, there is a matrix in which lookups can be done using product stock keeping unit (SKU). The resulting vectors are attribute vectors in which each column is a defined attribute and the value in a particular cell in the vector is the assigned quantified value based on the evaluation. The defined brand/label characteristic panel matrix is stored in the cloud storage and the data can be pulled down as needed. The data can be used in various use cases/modalities, as discussed in further detail below.
In disclosed embodiments, the data may be input into dynamic time warping (DTW) alignment calculations 208 which perform mathematical operations to determine how different the defined brand/label characteristic vectors are from one another. The results may be used to build a spatial ranking (SR) model, which is stored. DTW can be used to align vectors which are non-linear and widely distributed and collapse them into the same vector space. This includes calculating an energy cost to collapsing the vectors into the same space, i.e., a measure of how different the vectors are from one another. DTW allows the vectors to move in high-dimensional space, can be numerically expedited with algorithms to control the warping window, allows for the application of non-standard local weights, handles comparison vectors of differing length, and can be efficiently scaled with the increase of vector length.
It is possible to keep track, in high-dimensional space, of where the differences are most significant. This results in a raw value, i.e., a unitless value, which can be used to create a “heat map” of vector alignment costs. For example, such a map could have SKU values defining rows and columns. The distance, i.e., alignment cost, between two SKUs can be determined for a particular point in time. As an example, reading down a column of a particular SKU (e.g., a particular version of a wine or wine label) would give the range of the alignment cost to all other SKUs for the particular wine.
DTW algorithms are commonly used in finance and audio-visual processing. For example, it can be used to determine how different two waveforms are from one another. In disclosed embodiments, DTW calculations may be performed using a coded module, e.g., using Python, which includes the necessary mathematical functions to perform the calculations. In disclosed embodiments, a coding (e.g., script) language, such as Python, may be used to code the entire platform. DTW data models may be stored, for example, in a Python data format and the pre-computed models are stored in cloud storage.
The panels of defined brand/label characteristics can be used to build Gaussian Mixture Models (GMM) 212 which are a form of machine learning. A GMM model may be used to look at distributions of data and generate novel data that fits well within a given distribution. Such models can be used to create “algorithmic” products, e.g., algorithmic labels. Deep learning models 214 may be used to build neural networks to perform functions such as generating wine labels and performing high-dimensional product (e.g., wine) clustering.
In disclosed embodiments, the DTW model can be used in a DTW alignment use case when the business user is looking at a particular wine label, the website can present, e.g., the next five similar labels (not specific to the user, but specific to the wine being viewed currently).
In disclosed embodiments, a generative adversarial network (GAN) may be trained on classified data and later used to generate probabilities of features given a specific classification. If there were a sufficient number of labels and enough metrics to allow the algorithm to learn what makes a good label for a red wine, a white wine, a California red wine, etc., then the algorithm could generate label templates which could be used as the basis for a new label. Algorithms could also be used to determine what makes a “bad label,” e.g., a label which results in negative ratings during NB testing in which two bottles of the same wine are given different labels to evaluate the appeal of the label (i.e., consumers will rate bottles of wine which, unbeknownst to the consumer, are randomly labeled with Label A or Label B). For example, with enough learning, the algorithm could learn what font size is good for a particular flavor profile.
In disclosed embodiments, wine bottle labels can be used to build a data set, which might be filtered with respect to a particular region of interest. Additional metrics may be added, such as, for example, ratings, SKU performance, seasonal performance (e.g., winter versus summer label performance), and the chemistry of the wine. A generative adversarial network 306 may be used to allow the algorithm to make things, rather than just classifying things. The algorithm produces a new label template 308 by generating features of high probability, and as such may be a noisy template. The template may define specific characteristics of a desirable label, such as, for example, color (e.g., blue), font type, size, and color, label background texture, placement of text. Denoising 310 may be performed by a user, e.g., by shifting elements, sketching of the actual text to be used on the label, and performing label finalization. In some embodiments, denoising 310 may be performed by one or more NLP processes (alone or in combination with human denoising). One or more versions of the label may be produced and compared to existing labels that have not been doing well so market testing can be used, e.g., A/B testing. The resulting data can be fed back into the learning network to retrain 316 the algorithm. As more labels are made by this process, or by hand, the algorithm will develop a greater depth of knowledge of how to generate desirable wine labels. In disclosed embodiments, there may be feed forward generation of data, application of denoising, collection of market data (or actual production shipments of the labeled products to consumers and solicitation of feedback therefrom 314) to label the results, i.e., in the supervised machine learning sense, and return them as inputs to the algorithm for retraining 316.
In disclosed embodiments, a GMM model may be created based on a request by the wine producer for new SKUs (i.e., products). For example, it may be desirable to produce a new red wine label which will fit into the top ten red wine labels in a particular region. In such a case, a new “phantom wine label” 408 can be created which will be different from all of them (i.e., the top ten wine labels) but will have representative label characteristics of the top ten wine labels. The result provides a new panel of label characteristics, which can be used to produce a label template (e.g., a set of characteristics relating to font size, label background and text color, label style, etc.) which can be given to the business user as a basis for a new label design. The process, thus, works backward from the label characteristics and seeks to produce a wine label that has the desired characteristics.
In disclosed embodiments, there may be defined sets of wine labels of interest, e.g., top ten California red wine labels. A corresponding filter is applied to the panels of defined brand/label characteristics to pull in stylistic attributes. A GMM model may be built which samples the parameter space and provides a result which fits within the parameter space. The model generates a new feature vector of label characteristics, as noted above, which may be referred to as an “algorithmic wine label,” “phantom wine label,” or “virtual wine label.” This defines a wine label template which, if it existed, would have the characteristics that are representative of the applied set of criteria based on the wine labels of interest.
The phantom wine label 408 can be fed back to a design team to attempt to produce the hypothetical wine label. For example, there may be a number of label templates as a starting point, and the label designers try to create images, designs, text, etc., to match the label characteristics provided by the algorithm. The proximity to the target vector of label characteristics could be measured by application of DTW with the goal of achieving an alignment cost of zero. Samples of the wine label can be sent out at any step to obtain a label characteristic panel (e.g., a set of characteristics evaluated by an independent team of designers) to determine/judge the value of the various label parameters (including both objective and subjective criteria) such as traditional versus modern style, abstract versus figurative, bold versus subdued, so that adjustments can be made. In disclosed embodiments, a phantom wine label might be produced and used in a DTW alignment to known wine labels to obtain the next best choice relative to the phantom wine label. The label selected in this manner can then be used as a model/template for designing a new label for a different wine.
The server 530 is in communication with a label/indicator printer 540 which prints the labels on a roll of blank adhesive-backed labels in a continuous stream. The output of the printer 540 may also include indicator labels which perform particular functions, as discussed in further detail below. The stream of printed labels/indicators from the printer 540 is fed into a labeling machine, either in a continuous or batch manner. The labeling machine 550 receives an input of unlabeled, or partially labeled, filled wine bottles 560, attaches the appropriate label to each bottle 560 and outputs labeled bottles 570 to be shipped to the customers. As discussed in further detail below, the labeling machine 550 reads indicator labels included in the stream of printed labels from the printer 540 to determine how to match each label with the correct bottle of wine and the correct customer order.
In disclosed embodiments, a roll of blank labels feeds into the printer, e.g., a roll which is hundreds of feet long and stored on a separate unwinding unit. In disclosed embodiments, the roll of labels is unspooled and fed into the printer. The printed labels may pass directly from the output of the printer to a labeling machine (or subsystem) for application to the bottles. Alternatively, the printed labels may be stored on a winder and moved to the labeling machine/subsystem in batches.
In disclosed embodiments, the printed labels may be transferred between the printer and the labeling machine in batches. For example, the printed labels may be rolled up on a spool as they are printed. The continuous stream of labels may be cut at determined points to form batches. A batch indicator label may be used to determine the predetermined point at which the roll of labels is to be cut. The batch roll is then transferred to the labeling machine and set up to feed the labeling machine. The indicator labels of the batch may be read to determine a sequence of wine bottles to be placed on the conveyor for labeling. In disclosed embodiments, the labels are printed in the reverse order with respect to the order in which they are to be applied to the bottles because the first label read from the spool, i.e., unwound from the spool by the labeling machine, is the last label that was printed. The reading of the batch and indicator labels may be done using a bar code or other scannable code. The batch and indicator labels may also have the information in text form.
The stream of printed labels may also include order indicator labels and batch indicator labels between labels associated with specific customer orders. The order indicator labels provide information which, inter alia, makes it clear that a set of labels (and associated wine bottles) constitute a complete order and should go in one box together. The order indicator labels also make it clear that a specific number of bottles belong to a specific customer and provide the labeling sequence of the wines so that each label, i.e., label version, is attached to the proper bottle of wine. For example, the indicator (or “spacer”) label may have information indicating: “Customer ID=12345; Order=3 bottles; L1=a selected rosé wine, L2=a selected white wine; L3=a selected red wine.” The information may be encoded in the form of a bar code or other type of scannable code. In disclosed embodiments, the indicator label includes both printed information and a scannable code so that the information can be easily available to human handlers and machines. In disclosed embodiments, the warehouse for handling the bottled wine, before and after labeling, uses a barcode-based system, so barcodes related to the order on the indicator labels would be readable by the same system used to store and move bottled wine in the warehouse.
In disclosed embodiments, Customer Order A and Customer Order B may include the same three wines (i.e., identical brand and type) but the wines may have different labels for each customer. For example, Customer Order A may include a bottle of white wine with Label 2 and Customer Order B may include an identical bottle of white wine with Label 5. By virtue of the order indicator labels and batch indicator labels between labels associated with specific customer orders, the system can accurately provide label recipient correlation, i.e., a record of which customers have received which labels. Furthermore, the white wines may be identified with different SKUs, each of which can be associated with a specific version of a label, e.g., Label 2 or Label 5, to allow SKU-based distribution information to be accurately correlated with the particular labels on each product.
The indicator label helps to ensure that the order is correct, but also helps ensure that other orders are not subjected to errors. For example, if a wine bottle label were to be misprinted and become wrinkled and lost, a first customer's order of three bottles would receive only two correct labels and then the first label of a second customer's order might be applied to the third bottle. This could cause all of the remaining orders to be shifted by one bottle and, thus, incorrect. The indicator label helps to avoid this situation by calibrating the intervals between orders. As noted above, the indicator label also helps to ensure that the bottles and labels of each customer's order is properly aligned.
In disclosed embodiments, the back label 820 (see, e.g.,
As a practical matter, there are advantages to applying the back label before and separately from the application of the front label 810. Wine is typically made in large batches, e.g., in tanks, and then bottled. Moving the wine out of a winery in bottles is easier, from a regulatory point of view, if the bottles already have the legally required Surgeon General's warning (see
In disclosed embodiments, a high-scale facility, such as a winery, may be used to bottle large quantities of different types of wine to be labeled using the disclosed labeling system. In such a case, the filled bottles may be moved in bond (so that a back label is not necessary), which requires that the bottles are physically wrapped together in quantities for shipment and a barcode is applied to the outside of each shipping pallet. Bonded areas may be created in each of a number of distributed packing warehouses, and the shipped pallets of filled bottles may be unpacked in the bonded areas and labeled using the disclosed labeling system. In such a case, the labeling system would be set up within the bonded area, and the filled bottles would be labeled on the front and back at the same time. The bottles would be brought out of bond after labeling and would already be inside the packing warehouse ready for packing and shipment.
Filled bottles of wine having a back label already attached may be identified by that label, which may have a barcode. In some cases, the filled wine bottles may be easily identifiable, e.g., by color, if only one red wine, one white wine, and one rosé are offered. Otherwise, the cases of bottled wines are barcoded. In disclosed embodiments, each bottle is barcoded, each case of bottles is barcoded, and each pallet of cases is barcoded, each barcoded configuration having a separate SKU which identifies the type and quantity of the product.
In disclosed embodiments, a group of pallets, e.g., 10, 20, or 50 pallets, may contain a particular type of wine, such as red or rosé. The pallets may be stored in pallets racks. The pallets may be barcoded, as well as the corresponding pallet rack locations. The ordering subsystem could produce directions for a forklift driver to pick up a particular pallet for labeling. The operator would read the barcode on the pallet to verify that the pallet was in the correct position, and then would transport the pallet to the input of the appropriate (i.e., associated with the correct SKU) conveyor of filled wine bottles. The shrink wrap would be removed and the bottles would be manually or automatically loaded onto the conveyor for the specific wine type (i.e., for the specific SKU).
In disclosed embodiments, a quality control check of the labeled bottles may be performed to ensure that the correct labels have been properly applied to each bottle. This process may be made more efficient by providing an indicator, e.g., a small barcode in the bottom corner of each label, front and back, which could be optically read and checked in that manner.
In disclosed embodiments, the server detects that a label for a particular type of wine, e.g., a white wine, is being printed (or is about to be printed or has just been printed, depending upon the relative timing of the printer and the labeling machine) and sends a signal to a controller of the labeling machine (or directly controls the labeling machine) to release a bottle of white wine onto the labeling conveyor. In such an arrangement, instead of receiving a stream of labels, reading them, and then controlling the gate, the server will already have informed the labeling machine of the sequence of wine types which are coming up for labeling. For example, the server could indicate that, e.g., five bottles from now, a white wine bottle must be labeled. The controller of the labeling machine (or the server) controls the gate mechanisms based on a known or measured lag time between the release of a particular bottle from the conveyors of filled wine bottles and the actual labeling of the particular bottle. Thus, in disclosed embodiments, there are two different ways to operate the labeling machine: one in which printed labels are optically read as they are received by the subsystem; and one in which the server signals the labeling machine when to release a particular type of filled bottle of wine.
The labels may be characterized using a number of parameters, such as, for example, grape variety font type, grape variety font size, grape variety font color, grape variety vertical position, winery font type, winery font size, winery font color, winery vertical position, origin font type, origin font size, origin font color, origin vertical position, background color, etc. In this embodiment, the first label version (
In disclosed embodiments, an apparatus may be provided which comprises a general-purpose or special-purpose computing apparatus and which may execute program code to perform any of the functions described herein. The apparatus may comprise an implementation of one or more elements of the system 10, 100. The apparatus may include additional elements which are not mentioned below, according to some embodiments.
The above embodiments may be implemented in hardware, in a computer program executed by a processor, in firmware, or in a combination of the above. A computer program may be embodied on a computer readable medium, such as a storage medium or storage device. For example, a computer program may reside in random access memory (“RAM”), flash memory, read-only memory (“ROM”), erasable programmable read-only memory (“EPROM”), electrically erasable programmable read-only memory (“EEPROM”), registers, hard disk, a removable disk, a compact disk read-only memory (“CD-ROM”), or any other form of storage medium known in the art.
A storage medium may be coupled to the processor such that the processor may read information from, and write information to, the storage medium. In an alternative, the storage medium may be integral to the processor. The processor and the storage medium may reside in an application specific integrated circuit (“ASIC”). In an alternative, the processor and the storage medium may reside as discrete components. For example,
The computing system 1100 may include a computer system/server, which is operational with numerous other general purpose or special purpose computing system environments or configurations. Examples of well-known computing systems, environments, and/or configurations that may be suitable for use as computing system 800 include, but are not limited to, personal computer systems, server computer systems, thin clients, thick clients, hand-held or laptop devices, tablets, smart phones, databases, multiprocessor systems, microprocessor-based systems, set top boxes, programmable consumer electronics, network PCs, minicomputer systems, mainframe computer systems, distributed cloud computing environments, databases, and the like, which may include any of the above systems or devices, and the like. According to various embodiments described herein, the computing system 1100 may be a web server, a label data/test platform, or a labeling/distribution system (such as shown in
The computing system 1100 may be described in the general context of computer system-executable instructions, such as program modules, being executed by a computer system. Generally, program modules may include routines, programs, objects, components, logic, data structures, and so on that perform particular tasks or implement particular abstract data types. The computing system 1100 may be practiced in distributed cloud computing environments where tasks are performed by remote processing devices that are linked through a communications network. In a distributed cloud computing environment, program modules may be located in both local and remote computer system storage media including memory storage devices.
Referring to
The storage 1140 may include a variety of computer system readable media. Such media may be any available media that is accessible by computer system/server, and it may include both volatile and non-volatile media, removable and non-removable media. System memory, in one embodiment, implements the processes described herein. The system memory can include computer system readable media in the form of volatile memory, such as random-access memory (RAM) and/or cache memory. As another example, storage device 1140 can read and write to a non-removable, non-volatile magnetic media (not shown and typically called a “hard drive”). Although not shown, a magnetic disk drive for reading from and writing to a removable, non-volatile magnetic disk (e.g., a “floppy disk”), and an optical disk drive for reading from or writing to a removable, non-volatile optical disk such as a CD-ROM, DVD-ROM or other optical media can be provided. In such instances, each can be connected to the bus by one or more data media interfaces. As will be further depicted and described below, storage device 1140 may include at least one program product having a set (e.g., at least one) of program modules that are configured to carry out the functions of various embodiments of the application.
As will be appreciated by one skilled in the art upon reading this disclosure, aspects of the present application may be embodied as a system, method, or computer program product. Accordingly, aspects of the present application may take the form of an entirely hardware embodiment, an entirely software embodiment (including firmware, resident software, micro-code, etc.) or an embodiment combining software and hardware aspects that may all generally be referred to herein as a “circuit,” “module” or “system.” Furthermore, aspects of the present application may take the form of a computer program product embodied in one or more computer readable medium(s) having computer readable program code embodied thereon.
Although not shown, the computing system 1100 may also communicate with one or more external devices such as a keyboard, a pointing device, a display, etc.; one or more devices that enable a user to interact with computer system/server; and/or any devices (e.g., network card, modem, etc.) that enable computing system 1100 to communicate with one or more other computing devices. Such communication can occur via I/O interfaces. Further, computing system 1100 can communicate with one or more networks such as a local area network (LAN), a general wide area network (WAN), and/or a public network (e.g., the Internet) via network interface 1110. As depicted, network interface 1110 may also include a network adapter that communicates with the other components of computing system 1100 via a bus.
Although not shown, other hardware and/or software components could be used in conjunction with the computing system 1100. Examples include, but are not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data archival storage systems, etc.
According to various embodiments, the computing system 1100 may generate data that can be used to predict or determine which consumable (such as wine) a user may enjoy based on data associated with a known consumable. In general, the computing system 1100 may be used to implement any of the processes or analyses described elsewhere herein.
The foregoing diagrams represent logical architectures for describing processes according to some embodiments, and actual implementations may include more or different components arranged in other manners. Other topologies may be used in conjunction with other embodiments. Moreover, each system described herein may be implemented by any number of computing devices in communication with one another via any number of other public and/or private networks. Two or more of such computing devices of may be located remote from one another and may communicate with one another via any known manner of network(s) and/or a dedicated connection. Each computing device may comprise any number of hardware and/or software elements suitable to provide the functions described herein as well as any other functions. For example, any computing device used in an implementation of system 100 may include a processor to execute program code such that the computing device operates as described herein.
All systems and processes discussed herein may be embodied in program code stored on one or more computer-readable non-transitory media. Such media non-transitory media may include, for example, a fixed disk, a floppy disk, a CD-ROM, a DVD-ROM, a Flash drive, magnetic tape, and solid-state RAM or ROM storage units. Embodiments are therefore not limited to any specific combination of hardware and software.
Embodiments described herein are solely for the purpose of illustration. Those in the art will recognize other embodiments may be practiced with modifications and alterations to that described above.
This application claims the benefit of and priority to U.S. Provisional Patent Application Ser. No. 62/726,686 filed on Sep. 4, 2018, the contents of which are hereby incorporated by reference in their entirety for all purposes.
| Number | Name | Date | Kind |
|---|---|---|---|
| 6246326 | Wiklof | Jun 2001 | B1 |
| 8719347 | Tomkins | May 2014 | B1 |
| 20130080438 | Tompkins | Mar 2013 | A1 |
| 20160063595 | Oral | Mar 2016 | A1 |
| 20170103447 | Brandenberg | Apr 2017 | A1 |
| 20170236051 | van der Made | Aug 2017 | A1 |
| 20180047071 | Hsu | Feb 2018 | A1 |
| 20180315094 | Ashoori | Nov 2018 | A1 |
| Entry |
|---|
| Doucleff, Michaeleen. “Drinking With Your Eyes: How Wine Labels Trick Us Into Buying” (2013) NPR. (Year: 2013). |
| Boys, Callan. “Can you judge a wine by its label?” (2016) GoodFood.com (Year: 2016). |
| Bi, Hui et. al. “Accurate image segmentation using Gaussian mixture model with saliency map” Dec. 27, 2017. Pattern Analysis and Applications, pp. 869-878. (Year: 2017). |
| Sun, Xie et. al. “New Methods for Template Selection and Compression in Continuous Speech Recognition” Aug. 31, 2011. Interspeech, pp. 985-988. (Year: 2011). |
| Rasmussen, Carl. “The Infinite Gaussian Mixture Model” 2000. Advances in Neural Information Processing Systems, pp. 554-560. (Year: 2000). |
| Number | Date | Country | |
|---|---|---|---|
| 20200074517 A1 | Mar 2020 | US |
| Number | Date | Country | |
|---|---|---|---|
| 62726686 | Sep 2018 | US |
