Patent Application
20220043942
Numerous different three-dimensional (3D) printing processes exist. The idiosyncrasies of each different 3D printing process give rise to various challenges and opportunities for designers, engineers, and other users. In some instances, many users may design, modify, print or otherwise work with the same set or cohort of parts for 3D printing.
The written disclosure herein describes illustrative examples that are nonlimiting and non-exhaustive. Reference is made to certain of such illustrative examples that are depicted in the figures described below.
Designers, engineers, and other users of three-dimensional (3D) printers, such as multi jet fusion (MJF) printers, benefit by an understanding of the unique idiosyncrasies of the printing process they are using and/or object-specific considerations. For instance, different types of 3D printing processes may have different strengths and weaknesses. Process-dependent considerations may, for example, improve design efficiency and/or print quality outcomes. A system may provide process-specific recommendations and/or other notifications to a user during the design of a part or prior to printing a part. The notifications may also serve to continually teach and train users to design parts more efficiently and with greater skill. In addition, customized or curated learning material may be provided to the user, training the user in preparation for designing and printing new parts.
A system for selecting and prioritizing 3D part recommendations may analyze parts within a cohort of parts. The cohort of parts may be defined to include a set of related parts, a set of parts used by a defined group of people, a set of parts used within a specific industry, a set of parts for a particular machine or device, or other defined set of parts. The cohort of parts may be dynamically modified to include more or fewer parts over time.
The system may analyze the parts within the cohort of parts to identify notifications applicable to each respective part. The number of notifications applicable to a given part may, in some instances, be too numerous to reasonably present to the user without overwhelming the user. Accordingly, the system may select and prioritize notifications for display as part of a graphical user interface used for printing the 3D parts. In addition, customized learning material such as manuals, may be created for a user designing parts destined for the cohort of parts. In some instances, a new part may share similar characteristics or features with one or more existing parts. The learning material may include examples of notifications which occurred in these similar parts within the cohort of parts. The learning material may, in some examples, include similar parts from the cohort as teaching examples. In addition, the customized learning material may be different for novice and expert users. For example, a novice user may be taught issues which occur most-commonly among the cohort of parts, while an expert user, having already frequently seen the commonly-occurring cases, may benefit from training on the issues which have rarely or never been seen on parts in the cohort.
A notification count module 180 may determine a count value for each notification. The count value of a given notification is based on the number parts with which the notification is associated. A rare notification may, for example, be associated with a single part or just a few parts in the cohort of parts. A rare notification may have a relatively low count value. In contrast, a common notification may be associated with many or even all the parts in the cohort of parts. A common notification may have a relative high count value.
A first value module 182 may calculate a cohort-dependent recommendation value for each notification as an inverse function of the count value of each respective notification. Thus, relatively rare notifications with low count values are calculated to have a higher cohort-dependent recommendation value. Relatively common notifications with high count values are calculated to have lower cohort-dependent recommendation values.
A display module 184 renders a printable part to a user as part of a graphical user interface. The display module 184 may also identify notifications applicable to the printable part displayed as part of the graphical user interface. As previously noted, the number of notifications may be too great to present all of them to the user. Accordingly, the display module 184 may render at least some of the applicable notifications as part of the graphical user interface by filtering notifications as a function of their respective cohort-dependent recommendation values. The display module may order the notifications, for example, to show critical notifications first. The display module may display notifications as text or in other forms. For example, notifications may be in the form of text, drawings, 3D representations, audio, video, or the like.
In some examples, an optional time-dependent recommendation value module 186 may calculate a time-dependent recommendation value for each of the plurality of notifications. The time-dependent recommendation value may start at an initial value upon notification creation and decrease with time. Thus, the time-dependent recommendation value of each notification of each part may decrease with time. In such example, the display module 184 may render at least some of the applicable notifications as part of the graphical user interface filtered as a function of (a) their respective cohort-dependent recommendation values and (b) their respective time-dependent recommendation values.
In another example, an optional importance value module 188 may receive a user-defined importance value that is additionally used by the display module 184 to filter (e.g., select and/or order) which notifications are displayed to the user as part of the graphical user interface.
In still another example, a personalized recommendation value module 190 may associate a personalized recommendation value for each user of the system for each notification. The personalized recommendation value for a given notification may decrease from an initial value each time the notification is displayed to a given user. The display module 184 may render the applicable notifications as part of the graphical user interface filtered as a function of their respective cohort-dependent recommendation values and personalized recommendation values.
The display module 184 of the notification system 100 may render a select subset of the applicable notifications as part of the graphical user interface filtered as a function of their respective (a) cohort-dependent recommendation values, (b) importance values, and (c) personalized recommendation values.
In still another example, an experience value module 192 may associate an experience value based on their overall experience level with 3D printing and/or the user's experience using the specific printing process associated with the notification system 100, such as specific experience using MJF printers, selective laser sintering (SLS) printers, or another type of 3D print technology.
Examples of notifications that may be rendered for display to the user include warnings of potential print challenges, tips for increasing design efficiency, tips for increasing print resolution, recommendations for increasing the print speed of the part, possible errors, features that are too small to print accurately with a given printer, suggestions to improve print quality, and indications of suitability for three-dimensional printing. In some examples, different types of notifications may be displayed in color-coded format to facilitate differentiation thereof. For instance, fatal errors and serious warnings may be displayed in red, recommendations for improving the printing process (e.g., speed, resolution, quality, etc.) may be displayed in yellow, and notices that the part is well-suited for 3D printing may be displayed in green.
The notification system 100 presents a customized training experience in the form of selectively curated notifications based on one or more of the cohort-dependent recommendation values, the time-dependent recommendation values, the importance values, the personalized recommendation values, and/or the user experience values. The notifications may be process-dependent in that they relate to suggestions, warnings, and/or information arising from idiosyncrasies of the particular print process. In other examples, the notifications may alternatively or additionally be part-specific or user-specific and independent of the specific print process being utilized.
In some examples, the system may ensure that all users within a particular organization, team, company, industry, etc. are presented with the same notifications. Furthermore, the system may ensure that new notifications identified based on further or updated analysis of the parts in the cohort of parts are presented to each user associated with the cohort. Feedback on the relevancy and/or usefulness of notifications may be factored in to determine which notifications to present and/or the order in which they should be presented. Various machine-learning algorithms may be employed to adjust the importance values assigned to notifications for particular parts within the cohort of parts over time.
As previously noted, the cohort-dependent notification value can be calculated based on an inverse frequency function. In the following examples, a print recommendation, R, is used as an example of a type of notification. The cohort-dependent and frequency-dependent recommendation value, RCF, can be calculated using Equation 1 below:
R
CF=log(NP/(1+NR)) Equation 1
In Equation 1, NR is the number of unique parts viewed in the cohort for which the recommendation is made, and NP is the number of unique parts in the cohort.
A recommendation, or other notification, may be further associated with a recommendation importance value (or, more generally, a notification importance value). For example, a fatal error might be associated with a normalized importance value of 1, while insignificant informational messages might have a normalized importance value of 0.1 or even zero.
The newness of a recommendation may begin with a normalized initial value of 1 and decrease with time. A user-dependent newness of a particular recommendation for a particular part in the cohort of parts may initially be set at a normalized value of 1 and decrease each time the recommendation is displayed to the user or over time. In some examples, the user-dependent newness may decrease as a function of the number of times the specific recommendation is viewed relative to the total number of recommendations (or other notifications) the user has viewed during the same time period, as shown in Equation 2 below.
R
CN=log(NU/(1+NRU)) Equation 2
In Equation 2, RCN is the user-dependent newness, NRU is the number of unique parts viewed by the user for which the recommendation is displayed, and NU is the number of unique parts viewed by the user in the cohort. The system may combine the various values of each notification and render for display a subset of notifications having the highest total value. In some examples, a weighted function of the values may be utilized. In some examples, a recommendation or other notification may be displayed to a few people associated with a cohort of parts on a trial basis to determine the relevancy and helpfulness of the recommendation before presenting it to the larger group of users.
In the illustrated example, the filtered notifications 270 may include observations and recommendations concerning the model representation of the part 205. For example, the part 205 may have an unusually large number of triangles or an invalid mesh. The subset of notifications 270 may include recommendations and observations concerning the design of the part for printing. For example, if the part had walls that were relatively thin for use with a 3D printing process, a notification 270 may include an observation that the walls are thin and/or a recommendation to make the walls thicker.
The subset of notifications 270 may additionally include observations and/or recommendations concerning how to print the part with a particular print process. For example, with some print processes there may be a higher resolution in some axes, and so there may be an advantage in changing the orientation of the part when printing to take best advantage of the higher resolution.
Table 520 shows each part in the cohort and the notifications which may be shown for that part. The simplified example includes five parts associated with one or more of the three notifications. It is appreciated that a cohort may have tens, hundred, or thousands of parts and the number of possible notifications may be even greater. A training material module, implemented as software, firmware, and/or hardware may generate novice, advanced, and/or custom training manuals. A training manual may be a collection of organized notifications, example parts, and explanations of how each notification relates to the respective example part(s) for a cohort of parts. In other examples, the training manual may broadly include notifications, example parts, and explanations of how each notification relates to the respective example part(s) for multiple cohorts. The training manual may be presented as an integrated part of three-dimensional printing software, as a stand-alone software program, a document file (e.g., a PDF file), or even as a printed hard copy.
In the illustrated example, a novice training manual 530 includes a training material created for a novice user. The novice training manual 530 includes notifications which frequently occur, and which the novices are therefore likely to encounter. The novice training manual 530 also includes a detailed description related to each notification, and examples of parts from the cohort for which the notification might be shown.
The example parts may be selected to best illustrate the notification. For example, for notification R1, parts P1 and P5 are chosen ahead of part P2 because they only include that notification and so there may be less possibility of confusion. For notification R3, example part P3 is included because it is the only relevant part used by the cohort. Novice training manual 530 is further shown as including a part P100, which may not have been used in the cohort as an additional example to train on the notification R3.
Advanced training manual 540 shows training material for an expert user. The expert user is likely to have seen commonly-occurring notifications many times and so this advanced training can focus on notifications which occur less frequently. Advanced training manual 540 may also include detailed descriptions related to those notifications and example parts for which the notification would be shown. In addition, to further prepare the advanced user to design and print future parts, the advanced training manual 540 may include notifications, such as R200 which are not associated for any part in this cohort.
Various modules, systems, and subsystems are described herein as implementing one or more functions and/or performing one or more actions or steps. In many instances, modules, systems, and subsystems may be divided into sub-modules, subsystems, or even as sub-portions of subsystems. Modules, systems, and subsystems may be implemented in hardware, software, hardware, and/or combinations thereof.
Specific examples of the disclosure are described above and illustrated in the figures. It is, however, appreciated that many adaptations and modifications can be made to the specific configurations and components detailed above. In some cases, well-known features, structures, and/or operations are not shown or described in detail. Furthermore, the described features, structures, or operations may be combined in any suitable manner in one or more examples. It is also appreciated that the components of the examples as generally described, and as described in conjunction with the figures herein, can be arranged and designed in a wide variety of different configurations. Thus, all feasible permutations and combinations of examples are contemplated. Furthermore, it is appreciated that changes may be made to the details of the above-described examples without departing from the underlying principles thereof.
In the description above, various features are sometimes grouped together in a single example, figure, or description thereof for the purpose of streamlining the disclosure. This method of disclosure, however, is not to be interpreted as reflecting an intention that any claim now presented or presented in the future requires more features than those expressly recited in that claim. Rather, it is appreciated that inventive aspects lie in a combination of fewer than all features of any single foregoing disclosed example. The claims are hereby expressly incorporated into this Detailed Description, with each claim standing on its own as a separate example. This disclosure includes all permutations and combinations of the independent claims with their dependent claims.
| Filing Document | Filing Date | Country | Kind |
|---|---|---|---|
| PCT/US2019/029301 | 4/26/2019 | WO | 00 |
