Note : Les descriptions sont présentées dans la langue officielle dans laquelle elles ont été soumises.
CA 02588964 2007-05-18
WO 2006/054992 PCT/US2004/038621
AUTOMATED METHOD AND SYSTEM
FOR OBJECT CONFIGURATION
BACKGROUND
[0001] Designing a room requires numerous decisions to be made. For
each decision there may be many different choices. Such choices may
include room style (e.g., contemporary, classic, etc.), room size, room
layout, room color, price range, etc. Additional choices may include the
types or brands of products to be placed in a room (e.g., sink, range, etc.),
along with product size, color, material, price range, etc. The choices can
be based on many factors including consumer preferences, engineering
constraints, space constraints, industry standards, and design or aesthetic
considerations. Each decision made may constrain or otherwise influence
other decisions.
[0002] A"constraint" is broadly understood as a condition upon which
certain options may be selected. The set of options that could be selected
may be narrower after a given choice is made. On the other hand, it is
possible that the set of options could be broader once a given choice is
made. For example, if a four-foot cabinet is selected, the range of options
may be broader than if a five-foot cabinet is selected.
[0003] For example, with an "L-shaped" kitchen, a layout tivith the sink
in one leg of the "L" presents certain constraints: it may be desirable to
have a dishwasher next to the sink; or, a dishwasher and/or sink may only
be located adjacent to appropriate room-specific plumbing. Other
examples may include: a selection of a relatively small lcitchen, prevents
CA 02588964 2007-05-18
WO 2006/054992 PCT/US2004/038621
the use of a relatively large table; or selection of a kitchen in a
contemporary style precludes the selection of a sink in the "classic" style.
SUMMARY
[0004] An automated system and method is provided for configuring an
object (e.g., a room). In accordance with a preferred embodiment,
configuration is facilitated through the use of layering functionality and
frame-based inferences to evaluate stored knowledge of object attributes.
The frame-based inferences may be supplemented by a rules-based
inference system. In accordance with a preferred embodiment of the
invention, a graphics-based user interface may be used to permit
interactive configuration using two-dimensional and three-dimensional
models of the object, and incorporating engineering specifications, as well
as functional and physical attributes.
BRIEF DESCRIPTION OF THE DRAWINGS
[0005] FIG. 1 shows an example of a network arrangement in accordance
with a preferred embodiment of the invention;
[0006] Figs. 2-5 depict exemplary screen shots of a commercial
implementation of a preferred embodiment of the invention;
[0007] FIG. 6 illustrates layering functionality in accordance with a
preferred embodiment of the invention;
[0008] FIG. 7 illustrates user entry points in accordance with a preferred
embodiment of the invention;
2
CA 02588964 2007-05-18
WO 2006/054992 PCT/US2004/038621
[0009] FIG. 8 depicts an exemplary screen shot of a three-dimensional
model employed in a commercial implementation of a preferred
embodiment of the invention;
[0010] FIG. 9 depicts an exemplary screen shot of a photorealistic
rendering employed in a commercial implementation of a preferred
embodiment of the invention;
[0011] FIGS. 10 and 11 are illustrations of hierarchical structures used in
accordance with a preferred embodiment of the invention;
[0012] FIG. 12 depicts an exemplary screen shot of a bill of materials
employed in a commercial implementation of a preferred embodiment of
the invention;
[0013] FIG. 13 depicts an exemplary screen shot of a quote employed in
a commercial implementation of a preferred embodiment of the invention;
[0014] FIG. 14 depicts an exemplary screen shot of a shopping cart
employed in a commercial implementation of a preferred embodiment of
the invention; and
[0015] FIG. 15 depicts an exemplary sequence of options presented to
the user in configuring a room in accordance with a preferred embodiment
of the invention.
DETAILED DESCRIPTION
[0016] Preferred embodiments and applications of the invention will be
described herein. Other embodiments may be realized and structural or
logical changes may be made to the embodiments without departing from
3
CA 02588964 2007-05-18
WO 2006/054992 PCT/US2004/038621
the spirit or scope of the invention. Although the preferred embodiments
disclosed herein have been particularly described as applied to
configuration (e.g., a visual configuration) of an object such as a room, it
should be readily apparent that the invention may be embodied to provide
configuration functionality for any number of applications, services or the
like where configured and/or customized products or services are browsed,
selected, purchased, sold, supported or otherwise considered.
[0017] In accordance with a preferred embodiment of the invention, an
object configuration method and system is embodied in a single (or
multiple) processor-based system that may be supported in a stand-alone,
networked, mainframe, or client-server architecture. A single (or multiple)
program memory module is provided for storing one or more computer
programs used to perform the functionality described herein.
[0018] In accordance with a preferred embodiment, one or more user
interfaces are provided as part of (or in conjunction with) the object
configuration system of the invention to permit users to interact with the
system. Individual ones of a plurality of client devices (e.g.,
network/stand-alone computers, personal digital assistants (PDAs),
WebTV (or other Internet-only) terminals, set-top boxes, cellular/PCS
phones, screenphones, pagers, kiosks, or other known (wired or wireless)
communication devices, etc.) may similarly be used to execute one or more
computer programs (e.g., universal Internet browser programs, dedicated
interface programs, etc.) to allow users to interface with the configuration
system.
4
CA 02588964 2007-05-18
WO 2006/054992 PCT/US2004/038621
[0019] In accordance with a preferred embodiment of the invention, a
system in the form of a network arrangement is provided, as shown, for
example, in Fig. 1. As will be easily appreciated by one of skill in the art,
the arrangement can be modified to suit individual circumstances. In this
preferred embodiment, the system may be used to perform configuration
of an object such as a room. A user interface may be provided in the form
of a personal computer (PC) 100, which may be located; for example, at a
user's home, or in the form of an in-store computer or other processing
device such as kiosk 102 (or other mechanism in the store). In this
exemplary embodiment, the store may be a retailer that serves as the retail
outlet (e.g., sale, delivery, installation, etc.) for many or all of the
components used in the configured room. (It should be understood that
the term "component" comprises not only individual products but a
component can be a floor, wall, or any other aspect of a room used in room
design.)
[0020] A communications medium 104, which may take any form
(integrated, distributed, wired, wireless, etc.) such as the Internet, an
intranet, a local area network (LAN), etc., may be provided to provide
connectivity between a user interface and one or more programmed
computers (as symbolically represented by applications server 106 and
database 108) that execute the configuration functionality in accordance
with preferred embodiments of the invention. (It will be appreciated that
the configuration functionality as described herein may be located (in
whole or in part) in any other portion of the network (e.g., on PC 100, kiosk
102, etc.) in accordance with preferred embodiments of the invention.)
Further, the server 106 is in communication with database 108 in which
CA 02588964 2007-05-18
WO 2006/054992 PCT/US2004/038621
room configuration information is stored. Database 108 may be physically
or logically associated with server 106.
[0021] In accordance with a preferred embodiment, a user (e.g.,
consumer, sales-representative, buyer, seller, contractor, builder, architect,
consultant, organizer, project coordinator, etc.) interacts with the system to
configure an object such as a room. The interactive nature of the system
aids the user in arriving at the desired configuration including production
of any corresponding information (e.g., layouts, pricing, schematics,
product specifications, manufacturing requirements, parametric drawings,
etc.).
[0022] Figs. 2-5 depict an exemplary commercial implementation of a
preferred embodiment of the invention. Fig. 2 particularly depicts a screen
shot 400 accessible by a user through a user interface (e.g., PC 100, in-store
kiosk 102, etc.). In configuring an object in the form of a room, in this
implementation, the user is presented with several options. Those options
include a "Bathroom Design Model" option 402, a"What's on Display"
option 404, a "Bathroom Design Articles" option 406, a"Project Planner
option 408, a"What's New at Merchant" option 410 and a "Promotion and
Literature Request" option 412.
[0023] User selection of the "Bathroom Design Model" 402 option causes
the system to present a variety of different model bathrooms each having
different styles or other consumer applications, as illustrated in screen shot
500 (Fig. 3). Each of the different styles of the bathrooms may be pre-
configured (e.g., by the retailer) with different styles and arrangements of
appliances, fixtures, or other components (e.g., tubs, sinks, floor, wall
paper,
6
CA 02588964 2007-05-18
WO 2006/054992 PCT/US2004/038621
paint colors, windows, doors, etc.) to match the style of the room design.
For example, the bathroom style titled "Gentlemari s Retreat" 502 is pre-
configured with a plurality of components, colors and textures fitting a
traditional style bathroom having masculine features. In comparison, the
bathroom style titled "City Spa" 508 is pre-configured with an arrangement
of components different from those of the Gentleman's Retreat 502 and
which are relatively more contemporary in style. The information defining
the components of each of these exemplary bathroom styles and other
related data are preferably stored in one or more storage units accessible by
the system, as symbolically represented by database 108 (Fig. 1).
[0024] Referring now to Fig. 3, assuming that the user wishes to
configure the bathroom associated with the "Gentleman's Retreat" 502
model, the user selects the linked photograph 502, selects "Gentleman's
Retreat" 502 from menu 510, clicks on the image label, or otherwise
indicates the intention. As a result, the system retrieves the data
concerning the components making up the pre-configured model and
produces graphical images of the pre-configured components. These
graphical images can be viewed by a user (through a user interface, e.g., PC
100), as shown in Fig. 4. Fig. 4 is an exemplary graphical image in the form
of screen shot 600 that illustrates a 2-dimensional (2-D) image of a
Gentleman's Retreat bathroom model 602. This model bathroom is
characterized by a pre-selected arrangement of fixtures including sinks 604,
commode 606, tub 608, wall color 610, flooring 612, and other fixtures and
products.
7
CA 02588964 2007-05-18
WO 2006/054992 PCT/US2004/038621
[0025] Each component selected for bathroom model 602 (e.g., the tub
608) has various attributes stored in database 108. The tub 608, for example,
as used in room model 602, has specific dimensions, available colors,
material, and price. In keeping with the Gentleman's Retreat style, the
attributes of the tub 602 will be limited or constrained (i.e., limitations
placed on attributes that can be used in the Gentleman's Retreat style). For
example, the tub 602 may be limited to colors such as white and off-white,
the material of the tub 602 might be of a porcelain-type material, metal
material, or a polymer material, etc. These constraints imposed by the
selection of the Gentleman's Retreat style are typically pre-determined (e.g.,
for example, by the retailer) and may determine the set of component
attributes (or the components themselves) from which a user can choose to
include in the room to be configured. A given room style model may have
other configurable attributes related to the room. These attributes may
include size, shape, style, texture, etc. of the room or components within
the room, as well as cost figures (e.g., budget or price range of a given
product, project, etc.).
[0026] Constraints on values of the component attributes can be
provided directly by the user (or other entities), be obtained from database
108, derived from a combination of both, or otherwise provided as an input
to the system. For example, constraints can include space planning
constraints according to standards like the NKBA (National Kitchen and
Bath Association) rules, which, for example, specify where a sink might be
located in the room relative to a tub. Site-specific realities could be taken
into consideration as well (e.g., no piping in a region of a given bathroom
precluding the building of a sink in that region).
8
CA 02588964 2007-05-18
WO 2006/054992 PCT/US2004/038621
[0027] It will be appreciated that the constraints may be implemented
from a spatial point of view as one or more of point-to-point constraints
(e.g., an immobile camera as the perspective from which the pictorial view
is rendered), point-to-line constraints (e.g., where piping may be located),
or point-to-plane restraints (e.g., where a mirror is located on a wall).
Further, constraints may apply to 2-D but not 3-D models, to 3-D but not 2-
D models, etc.
[0028] A retailer may recommend certain products which default as the
first choices for a given room style (e.g., Gentlemaiz s Retreat) selected by
the user. This is beneficial for a retailer that wishes to promote one or more
brands. While only one product may be recommended by the retailer,
there may very well be several others that also fit the same description and
fit within the user-specific constraints.
[0029] In accordance with an exemplary embodiment of the invention,
although a given room design is pre-configured with a set of recommended
components for each given room style, the user, e.g., the consumer
browsing the system at home on a PC, may further customize the
components (e.g., selection, arrangement, etc.) as used in the room to be
configured. For example, to customize the bathroom model selected, the
user may wish to change the flooring 612. To accomplish this, the user
selects the flooring attribute and modifies it by selecting the flooring 612
on
the'display screen or employing a menu 614 that enables input of the
.flooring alternative (e.g., change floor color, type, material, etc.). Upon
modification of the component attribute, the applications server (106 of Fig.
1) has reconfigured the room according to the desired preference. Once the
9
CA 02588964 2007-05-18
WO 2006/054992 PCT/US2004/038621
flooring attribute is changed, the configuration will include the new
flooring attribute and the system will output an updated graphical image
showing the configuration with the new flooring attribute (e.g., 706, as seen
in Fig. 5). As shown between screen shots of Figs. 4 and 5, the original tile
flooring 612 (Fig. 4) with a diamond-shaped pattern on a light background
is changed (Fig. 5) to flooring 706 with a wood-grained appearance.
[0030] Of course, other attributes of any component of the room model
602 may be changed by the user. For example, the user may choose to
modify the tub 608 style. This would be accomplished by selecting the tub
608 and choosing from a plurality of other tub styles stored in database 108.
The same can be done for any component in the room where the list of
available components and/or attributes depends on pre-selected constraints
(e.g., which have been imposed by the retailer, the user, or other). For
example, in this exemplary embodiment, the, retailer has pre-determined
the attributes of other tubs that would be suitable for use in the
Gentleman's Retreat style room model 602 aside from the one
recommended tub 608. The list of other available tubs may also be
shortened or lengthened depending upon other data entered by the user
(e.g., budgeting constraints, room size constraints, etc.).
[0031] Figs. 4 and 5 also illustrate the resulting product of a feature of a
preferred embodiment of the invention referred to as "layering
functionality." Layering functionality simplifies and makes more efficient
the processing of numerous high-quality images as the user makes
selections in real-time. Where each component in a room could be selected
from a plurality of different alternatives, multiple unique images could be
CA 02588964 2007-05-18
WO 2006/054992 PCT/US2004/038621
generated and stored to present the user with all possible selections. For
example, in an exemplary scenario, where the wall color could be selected
from 100 different colors, cabinets selected from 10 different types of
cabinets, and flooring selected from 10 different types of floors, 10,000
unique images would have to be generated and stored. In accordance with
a system employing the layering functionality in accordance with a
preferred embodiment, however, an image of the room can be divided into
separate layers (e.g., three layers: wall colors, cabinet type, and floor
style)
such that only 121 images (including the original image) need to be stored
under the same scenario.
[0032] Generating these layers can be accomplished by modeling three-
dimensional (3-D) information onto a 2-D image. That is, an original 2-D
image of, for example, a room, is processed by the automated system
designer manually (or automatically by the system) such that 3-D reference
data is added to the data of the original 2-D image. The 3-D information so
modeled can include location of the camera, location and intensity of light
sources in the picture, identification of various planes (i.e., floor, counter-
level, cabinet bottoms, cabinet tops intersecting walls, roof, floor, etc.), 3-
D
identification of components like counter-tops, cabinets, etc., and three-
dimensional visual behavior data of components like cabinet doors. The 3-
D data added to the 2-D image may be stored, for example, in a
commercially available image rendering software package. This 3-D
information is used by 3-D rendering tools to separate the picture into
layers 200, 202 and 204, as seen in Fig. 6.
11
CA 02588964 2007-05-18
WO 2006/054992 PCT/US2004/038621
[0033] In a preferred embodiment, the system generates individual
layers, each exclusively depicting at least one component (e.g., floor, wall,
cabinet, etc.) on a single layer. The generation of such layers may be
accomplished with a commercially available image processing package
such as, for example, Adobe Photoshop, etc. The single layer presents the
depicted component with its recommended (or customized) attributes (e.g.,
size, proportion, spatial arrangement, etc.), while the remaining portion of
the layer is made transparent. In the exemplary scenario discussed above,
the system could be used to generate three new images in addition to the
original image (as shown in Fig. 6): a floor layer (200) that depicts only the
floor; a wall layer (202) with only the walls showing; and a cabinet layer
(204) with only cabinets showing.
[0034] Next, each layer is processed, or re-rendered, by a rendering
engine. In accordance with a preferred embodiment, a commercially
available engine may be used such as, for example, Adobe Photoshop,
GIMP, etc. Preferably, the rendering engine takes each 2-D layer image
along with the 3-D information and generates all the required variants of
the layer. For example, all 100 variants of the wall color layer are
generated and stored prior to the user requesting that color. The same
process is used to pre-generate all 10 cabinet styles and all 10 floor styles,
etc. (This process may be performed by the software in batch offline, so
that the variants of the layers can be generated with very high quality.) As
previously mentioned, the total number of images generated is 100 + 10 +
+ 1 = 121, rather than 10,000.
12
CA 02588964 2007-05-18
WO 2006/054992 PCT/US2004/038621
[0035] The 121 images are then indexed and stored. The index to these
images stores a sequence in which these images need to be superimposed.
The sequence of superimposition and the correct image set from the library
of 121 pictures is thus identified and a unique sequence of superimposed
images for each of the 10,000 possible combinations is generated.
[0036] It will be readily appreciated by those of skill in the art that each
layer (e.g., wall, floor, commode, tub, etc.) can comprise multiple attributes
(e.g., color, texture, materials, finish, etc.).
[0037] In accordance with a preferred embodiment of the invention,
visual attributes of a given room are also linked to dynamic catalog data.
For example, in accordance with a preferred embodiment, a database 108 is
compiled comprising products available for use in room configuration.
The database may include product information from many different
manufacturers including product color, size, compatibility with other
products, pricing, etc. That is, the product parameters found in one or
more manufacturer's catalogs are included in the database. For example,
when a user selects a white tub, there may be several different white tubs
(supplied by several manufacturers) in the database 108. However, based
on the applicable constraints, only a subset of those white tubs may be
suitable for use in the user-selected room style (e.g., Gentleman's Retreat),
thus narrowing the field of suitable tubs.
[0038] In accordance with a preferred embodiment of the invention, the
system provides a plurality of entry points by which a user can undertake
room configuration. An exemplary implementation of a plurality of entry
points is represented by entry points 300 in Fig. 7. Entry point 302
13
CA 02588964 2007-05-18
WO 2006/054992 PCT/US2004/038621
represents the above-described process in which the customer preferably
selects a room (e.g., bathroom or kitchen), room type (e.g., master bath or
guest bath), room style (e.g., traditional or contemporary), and/or a room
layout. The user will then be presented with a suggested room design,
populated with matching fixtures and other components.
[0039] Another entry point 304 represents the presentation of a room
based on an advertisement or in-store display. At this entry point, the
customer can select from room models that are pre-configured as in an
advertisement (e.g., magazine, television, Internet, etc.), in-store display,
or
other promotional presentation.
[0040] In entry point 306, the customer can design his or her own room
from "scratch." This may also be considered "expert mode." The customer
can select any components in database 108 (or other sources) for use as
specified to create completely customized rooms.
[0041] After any of the entry points 302, 304, 306, a user has an
opportunity (at point 308) to modify one or more components or other
aspects of the configured room.
[0042] In accordance with a preferred embodiment of the invention,
carrying out object configuration may.utilize an inheritance-based
approach to modeling the data. A room, for example, is defined by certain
attributes, including: wall colors and textures; floor types, colors and
textures; cabinet types and colors; counter top types and colors; and fixture
types and colors. A room can be customized or otherwise specialized. A
kitchen may have a stove type, dishwasher type and refrigerator type as
14
CA 02588964 2007-05-18
WO 2006/054992 PCT/US2004/038621
additional descriptors. A bathroom has a sink type, bathtub type, faucet
type, etc. The kitchen further can be specialized into L-Shaped, galley, etc.
The bathroom specializes into master, powder, etc. Each level of
specialization presents certain attribute constraints. In accordance with an
embodiment of the invention, each of the user-selectable attributes are
further linked to additional attributes such as engineering specifications of
the various manufacturers' products (size, capacity, etc.) that fit into the
specific room.
[0043] In accordance with an embodiment of the invention, the system
guides the user via a series of questions to select the attributes (e.g.,
visual
attributes) of the object (e.g., room) being designed or configured. These
choices change the look of the object and at the same time are dynamically
linked to specifications, engineering and other product data of the various
different manufacturers' products or components that are needed to make
the object.
[0044] In accordance with a preferred embodiment of the invention, at
any time during the object configuration, the user is able to view a 3-D
rendering of the object as configured so far. Fig. 8, for example, illustrates
a screen shot 800 that represents an exemplary rendering 802 of a room
during the configuration operation. Using this rendering during the
configuration, the user is able to benefit from a pictorial rendering of the
relationship in order to visualize the room so configured. The 3-D model
presents a view that enables the user to gain different perspectives inside
tl ie room and to visualize better how the room will look.
CA 02588964 2007-05-18
WO 2006/054992 PCT/US2004/038621
[0045] In accordance with a preferred embodiment of the invention, the
3-D view is navigable. That is, the user is able to navigate through the
view to approach or retreat from the room, "walk" through the room to see
it from a different angle, etc. The 3-D viewer used may be one that is
generally commercially available, such as one developed by View 22
Technology, Inc,
[0046] In accordance with an embodiment of the invention, while
viewing the 3-D rendering of the room, space planning decisions may be
integrated with product feature decisions. As a result, the user is able to
move components about the room, add components, delete components,
etc. The system identifies a 3-D model based on engineering and other
attributes. These 3-D models are then available to interact with a three-
dimensional software representation of a room and associated space
planning constraints, including but not limited to the NKBA (National
Kitchen and Bath Association) rules.
[0047] In accordance with a preferred embodiment, the system can
model product catalogs that link product attributes to three-dimensional
models. In the embodiment, the linking can be accomplished by modeling
the product attribute data in a hierarchy of specialization layers. These
layers of specialization are arranged in a tree structure. Each node of the
tree represents data that effectively overwrites or extends the attributes of
the product group it specializes (the parent node). A node, for example,
can be used to overwrite the dimensions and the three-dimensional
component features. In an exemplary operation, the user may start with
the engineering attributes deduced from the visual decisions made at the
16
CA 02588964 2007-05-18
WO 2006/054992 PCT/US2004/038621
room design session. The system then helps the user pick more of the
erigineering attributes that continuously overwrite the most appropriate
three-dimensional model. The three-dimensional model is then introduced
into a three-dimensional representation of a room with space-related
constraints.
[0048] In a preferred embodiment of the invention, the user is able to
view a high-quality photograph-based image 902 of the room so re-
configured, as shown in Fig. 9, on screen shot 900. The image enables even
greater detail and realism, permitting the custom-configured bathroom to
be seen virtually as it will be in the actual bathroom itself. As can be seen,
the locations of the objects within the room and the number of objects in
the room differ from that of Figs. 4, 5 and 8.
[0049] In accordance with a preferred embodiment of the invention,
whether graphics are employed or not, the configured room is a construct
that may take into account many attributes and interrelationships and
dependencies thereof. It may take into account relationships between and
among attributes of the components making up the room, attributes of the
space defining the area in which the components are to be placed, and the
relationship of multiple components.
[0050] In accordance with a preferred embodiment of the invention, an
object hierarchy or inheritance (e.g., frame-based) functionality may be
used to process knowledge (e.g., data concerning components, space, etc.)
used in the system, which may be supplemented by an interpreted rules
system. The components may be, for example, products such as room
fixtures, and the space may be a given room. In a typical example, the
17
CA 02588964 2007-05-18
WO 2006/054992 PCT/US2004/038621
components and the room have specified attributes. By way of non-
limiting example, component attributes include but are not limited to type
of product, size, color, material, price range, etc. And, by way of non-
limiting example, room attributes may include but are not limited to room
size, layout, style and price range.
[0051] In accordance with a preferred embodiment, user data is input
through user input of answers to a series of questions regarding
configuration of objects. When given the value of one or more answers, the
system infers the values of answers to other questions automatically, and
thus eliminates the need for excessive rule constructs, as typically required
in a rules-based system. Through inference, the system may also remove
or insert questions (and their associated answers) based on the user's
previous response(s).
[0052] As the artisan will appreciate, attribute-based functionality can be
obtained by using an attribute inheritance engine (sometimes referred to
generically as a frame-engine). However, although one of skill in the art
will understand how to implement the functionality in accordance with the
invention, the precise nature of the modules as programmed may differ in
light of individual circumstances, taking into account, for example,
functional specifications that need to be met, the programmer's style, and
interoperability with other programs.
[0053] In accordance with a preferred embodiment of the invention, the
system is primarily constructed using an object hierarchy or inheritance
(e.g., an attribute-based) knowledge organization, representation, and
18
CA 02588964 2007-05-18
WO 2006/054992 PCT/US2004/038621
classification. Attribute-based and rule-based expert systems encode
knowledge using fundamentally different models.
[0054] In the context of product configuration, the problem an expert
system is supposed to solve is as follows: given a set of options ("questions"
to be presented to the user), each of which has a set of legal attributes
("values" with which the user may answer the questions), how are the
options/attributes themselves related to each other (which
options/attributes remove [or "constrain away"] other options/attributes,
which options/attributes are "interdependent", etc.).
[0055] As an example, if the user has answered options A, B, and C with
attributes x, y, and z, the expert system must then determine what are now
the legal attributes for some other option D. In order to do this, an expert
system first has to encode the relationship between the options in some
manner (i.e., represent the knowledge). Then, as the expert system is
provided current choices as inputs in real time (i.e., "the user has answered
options A, B, and C with attributes x, y, and z"), the system must now
apply its encoded knowledge to the problem "what are the legal attributes
for D" in order to supply the solution(s).
[0056] A rule-based expert system generally represents knowledge using
a large collection of If-Then-Else constructs (i.e., "If the user has answered
option 'A' with attribute 'x', Then attribute 'y' is no longer available for
option 'B', Else . . . "). This approach, however, is a unidirectional
encoding: if the conditional portion (the user has answered option "A"
with attribute "x") of the statement is true, the resultant portion must also
be true (attribute "y" is no longer available for option "B"). The converse
19
CA 02588964 2007-05-18
WO 2006/054992 PCT/US2004/038621
(if attribute "y" is no longer available for option "B", then the user has
answered option "A" with "x") is not necessarily true, nor in general
should it be.
[0057] However, a rule-based expert system requires that the inverse
rule must be written separately if it is deemed necessary. If the product
knowledge beiing modeled is complex, the required rules and inverse rules
can grow into a very large number. In addition, applying this type of
knowledge representation to the problem "the user has answered options A,
B, C with attributes x, y, z; now what are the legal attributes for this other
option D" involves using an interpreter or compiler that understands the
rule syntax so as to apply the rules to the problem.
[0058] On the other hand, the attribute-based functionality as used in a
preferred embodiment of the invention encodes knowledge in an entirely
different manner that does not suffer from this "inverse rule" problem. The
knowledge is represented in the object hierarchy in a hierarchical tree-like
structure. The nodes of the tree are generically called "frames" (e.g.,
corresponding to product categories) and each node contains a collection of
"slots" (e.g., corresponding to product features and options). A slot is a
one-to-many relationship between an option and a subset of the legal
values (e.g., attributes) for that option.
[0059] Textually, a slot is represented as follows:
A=x, y, z
where "A" is an option, and "x", "y', and "z" are attributes for
option A.
CA 02588964 2007-05-18
WO 2006/054992 PCT/US2004/038621
[0060] In general, in the type of object hierarchy known as frame/slot
hierarchy, a frame contains multiple slots, and has child frames as well
(because of the tree structure). Conceptually, all the slots in a frame "go
together." If at least one of the slots in a frame is found to be invalid
(e.g.,
Slot "A=x, y, z" is in a particular frame, but the user has answered "A" with
attribute "w"), then the entire frame (along with all of its other slots and
all
of its child frames) is invalid. Functionally, products and attributes in such
a frame are removed from the configuration process. Applying this
knowledge representation to the problem "the user has answered options A,
B, and C with attributes x, y, and z -- now what are the legal attributes for
this other option D" involves first marking the appropriate portions of the
tree invalid as the user supplies answers to options. Then, to actually
provide an answer to the question "what are the legal attributes for some
other option D" the features will look for all the frames which have not
been marked invalid and which reference "D" (i.e., have a slot involving
"D") and then combine all the attributes found in these slots (eliminating
any duplicates). At the conclusion of this process, the answer (in many
cases, a multi-faceted answer in terms of associated data [a price, an image,
a set of drawings, etc.]) is generated and displayed.
[0061] In accordance with a preferred embodiment of the invention,
incorporating component knowledge into the frame-based system
enhances its operation.
[0062] The following simple example will illustrate the differences
between an If-Then-Else rule based system and one utilizing frame-based
functionality in accordance with a preferred embodiment of the invention.
21
CA 02588964 2007-05-18
WO 2006/054992 PCT/US2004/038621
[0063] Assume there are four people who are identified by a letter, color,
and number.
Jim is A, Blue, and 3.
Ted is B, Red, and 2.
Randy is A, Red, and 4.
Roy is B, Blue, and 2.
[0064] The first screen in an application would prompt the user for
choosing a letter. (As an additional requirement, the questions can be
answered in any order, and the user might skip the first screen and come
back to it later.) The If-Then-Else rules to handle such option/attribute
pairs are as follows:
if Color has no answer and Number has no answer then
Letter is A or B
else if Color has no answer then
if Number is 3 or Number is 4 then
Letter is A
if Number is 2 then
Letter is B
else if Number has no answer then
Letter is A or B
else if Color is Blue and Number is 3 then
Letter is A
else if Color is Red and Number is 2 then
Letter is B
else if Color is Red and Number is 4 then
Letter is A
else if Color is Blue and Number is 2 then
Letter is B
[0065] Two more sets of rules will still have to be written for the Color
screen and the Number screen. The difficulty of adding a new person to the
22
CA 02588964 2007-05-18
WO 2006/054992 PCT/US2004/038621
data or adding another class of question to determine the group's favorite
fruit can easily be seen.
[0066] In contrast, the frame based functionality utilized by the
invention permits the entry of that same data as follows. After adding the
four questions to the system, add the following compound slot:
LETTER COLOR NUMBER PERSON
A Blue 3 Jim
B Red 2 Ted
A Red 4 Randy
B Blue 2 Roy
[0067] To add a new person, a new row is simply inserted, adding the
new person's letter, color, number, and name. To add another question
such as the group's favorite fruit, a new column is simply added to the slot,
and a list of everyone's favorite fruit can be added.
[0068] While the actual algorithms as implemented are likely to be much
more complex, the following illustration may be useful:
[0069] In the beginning of a configuration session all frames are valid,
and the example above has four frames, one for each row. When asked for
the list of available answers for any question, the system looks down the
column for that question. If the row is valid, its answer is added.
Accordingly, for the Letter question, the system sees A, B, A, B. The
23
CA 02588964 2007-05-18
WO 2006/054992 PCT/US2004/038621
duplicate answers are combined to arrive at A, B. Now suppose the user
answers A. This means that unless the frame has Letter=A, it is invalid.
The second and fourth rows are thus invalid. Next the user is presented
with the Color question. To find the answers, search down the Color
column; the result is Blue and Red for the available answers (Blue from the
first row and Red from the third). The user picks Blue. This choice makes
the second and third rows invalid.
[0070] As a result, there is only one valid row left, the first row. The
user has effectively finished the selection process by answering only two of
the three questions.
[0071] Frame-based operation is particularly useful when applied to
real-world, complex product knowledge challenges. Real-world product
knowledge contains relationships between products, knowledge common
among similar products, knowledge common among different products,
and exceptions to all of the above.
[0072] An important relationship among products can be expressed as:
"Product A is a type of Product B" (e.g., a casement is a type of window, a
,
car is a type of vehicle). This relationship is called inheritance.
Inheritance
is a parent to child relationship, but not in the traditional sense. In human
beings, if a parent has a trait, the child may or may not share that trait
(e.g.,
brown hair). In this form of knowledge inheritance, the child must inherit
all traits. Inheritance is important, because it allows the software engineer
to combine all the identical traits for the children in one logical place: the
parent. Each frame is a parent with children, which in turn have their own
children. For example, if there were a system for selecting groceries, the
24
CA 02588964 2007-05-18
WO 2006/054992 PCT/US2004/038621
logical way to organize the data would look something like that shown in
FIG. 10. The frame based system allows the construction of this tree with
the result illustrated in FIG. 11.
[0073] As illustrated, when the Bread frame is selected, there is a trait
defined as bread. This means that all types of buns and loaves can be made
out of White, Wheat, or Rye bread, because both Buns and Loave& inherit
from Bread. If any special types of Buns were defined by adding "children"
to the Buns frame (i.e., hamburger or hot dog) these new types also inherit
White, Wheat, or Rye. Children inherit everything from their parents,
grandparents, and great-grandparents, all the way up the tree. In the
future, if the store added Pumpernickel and carried it for all types of bread,
such information could be added to the Bread frame. If the store offered
Pumpernickel for Loaves only and not Buns, then Pumpernickel could be
added to the Loaves frame. Without the power of inheritance,
Pumpernickel would have to be manually added to every type of bread. '
[0074] There are some types of product knowledge that may be shared
among different frames in the tree, but which cannot use inheritance. In the
groceries example, a trait of this sort would be packaging. Not all groceries
come in packages, and sometimes the same kind of product might be
available with or without packaging (for example, packaged bread versus
bread from the deli). If packaging is available, there will be some product
knowledge that applies to all packaging (e.g., servings per container). A
frame can be defined by itself to represent such knowledge, which can then
be added into the. tree at whatever points are appropriate.
CA 02588964 2007-05-18
WO 2006/054992 PCT/US2004/038621
[0075] In accordance with a preferred embodiment, although frame-
based inference is used, the system also includes the ability to process data
using rules files, enabling the approach to be more of a hybrid as desired.
It is foreseeable that some custom answers are more conveniently handled
using rules files rather than building the product knowledge. A rules-
based filtering methodology may be used, for example, for filtering output
from the frame-based functionality to comply with certain rules established
for a particular product, product attributes, room attributes, or a
combination thereof. Similar filtering of unwanted answers, or
inapplicable questions may be useful. Rules-based functionality may also
be used to add special answers, perform calculations, generate user
warnings, or any other special processing required for specific system
implementations. Rules-based functionality may be particularly suited for
computing the price of a room and distribution of any discounts that are
available to the room.
[0076] In accordance with a preferred embodiment, rules-based
functionality may be added to supplement the operation of frame-based
functionality. Rules-based functionality is typically employed to handle
special cases, exceptions, and functionality that are specific to a product
line or product series.
[0077] In accordance with a preferred embodiment, any number of
additional modules may be added to serve a supportive and optional role
(called on an "as-needed" basis). Such modules may relate to pricing. For
example, price reports can be graphically generated and prices can be
calculated and displayed to the user in multiple currency types. Further, a
26
CA 02588964 2007-05-18
WO 2006/054992 PCT/US2004/038621
CAD (Computer Aided Design) module may be added to read CAD
drawings and enable multiple CAD drawings to be displayed
simultaneously, with separate components to be overlaid upon one another
to form a complex illustration. Specifically, the CAD functionality may
handle the determination of which components need to be overlaid and
may present a list of the files containing the required components to a CAD
control, which may in turn read the CAD files and display the components
on the user interface. The CAD functionality may also print, copy or
otherwise output the CAD files.
[0078] In addition, instead of inputting user data (e.g., room or house
dimensions) from scratch, the user may initiate access to an input file (e.g.,
CAD file) with such data.
[0079] Further, the system may be programmed to output files (e.g.,
CAD files) once all components have been specified and located in a room.
In this manner, a contractor can effectively work with the object as
configured by the customer.
[0080] A preferences module may also be included in the system. A
preferences module allows the user to define a set of preferred answers for
questions applicable to the design choices. In a preferred embodiment, the
set of preferred answers may be based on user-selected preferences,
regional specification preferences, manufacturer compatibility preferences,
etc. As an example, the user can initiate a configuration session and choose
a desired fixture brand, such as Kohler. The user can then set Kohler as a
preference for that entire project. As a result, in considering the object
attributes, the system will reference the preferences module and will
27
CA 02588964 2007-05-18
WO 2006/054992 PCT/US2004/038621
automatically retrieve the preference values by default and override other
items. The preferences module can also automatically inform a user that
the selected preference is unavailable for a given product during the
configuration of that product.
[0081] The above features can further be made part of a turnkey home-
design solution by enabling the user to view a bill of materials and pricing
range for the customized project (as shown in Fig. 12), generate a quote for
the bill of materials (as shown in Fig. 13), and convert the configuration
project into a shopping cart view and execute the purchase after changing
any product attributes from the detailed product database (as seen in Fig.
14).
[0082] The user is able to view a bill of materials and pricing range for a
particular project. Fig. 12 illustrates a screen shot 1000 of an exemplary
implementation of a preferred embodiment of the invention. As shown,
the products 1002 will populate the room. A quantity 1004 is shown for
each product. A model number 1006 is given for each product. Also, a
description 1008 is given for the product. If appropriate, a color/finish 1010
is given. Further, a price range 1012 is supplied. Additionally, the user has
the option to save the room 1014. By saving the room, the attributes
selected are saved for later use. Even more, the user has the option to
email the room 1016 so configured to another user. The other user may be,
for example, a family member for his or her review. Alternatively, the
other user may be a merchant who is able to carry out the purchasing and
installation of the room so configured, or give advice.
28
CA 02588964 2007-05-18
WO 2006/054992 PCT/US2004/038621
[0083] In accordance with a preferred embodiment, a quote for the bill of
materials may be generated. Fig. 13 illustrates a screen shot 1100 of an
exemplary implementation of a preferred embodiment of the invention.
The screen shot 1100 presents: a date 1102, project number 1104, customer
name 1106, customer phone 1108, and customer address 1110, as well as a
project name 1112 and email address 1114 to which the project can be sent.
A graphic representation of the configured room 1116 can be presented,
along with product code descriptions 1118 and price ranges 1120 therefor,
leading to a total price 1122 calculated from the price ranges 1120. If the
user so desires, the screen may suggest 1124 that the user print out the
order and give it to a sales associate to purchase the components of the
room. The user is also able to return to the quotations given (e.g., as in
Fig.
12).
[0084] Preferred embodiments of the invention may allow modeling a
"has a" relationship between rooms and the components therein. An
implemented room configuration system, for example, can refine a list of
products (e.g., bill of materials) in a customized room as the attributes of a
room change. As an illustration, a room with red walls, silver appliances
and wood floor can have GE appliances with model numbers #1234 and
#2345R, and red color paint and wood flooring. A different incarnation of a
similar room may have Whirlpool #9876 and WPL #5432 appliances, #345
wallpaper and @Asv floor tile from Daltile. As the user modifies the room
desired, the "has a" relationships can differ as well.
[0085] The user also has the option to consummate the purchase
electronically. The user can convert the project into a shopping cart-type
29
CA 02588964 2007-05-18
WO 2006/054992 PCT/US2004/038621
view, as shown by the exemplary implementation of a preferred
embodiment illustrated in Fig. 14. As shown in the exemplary screen shot
1200, the shopping cart contains data relating to the customized room
including, for example, a quote ID 1202, customer 1204, and room
description 1206. Further, the shopping cart can contain the item
designation 1208, the quantity/description thereof 1210, manufacturer 1212,
estimated delivery date 1214 and total price 1216. The user is able to delete
1218 a product from the shopping cart, copy 1220 the product (for example,
the user wishes to have two identical sinks), or continue with room design
1222.
[0086] Further, from the shopping cart view depicted in Fig. 14, the user
may refer back to the 3-D representation (e.g., of Fig. 8) and/or view a plan
view of the configured room with the specific appliances now to be finally
selected prior to placing the order. In this manner, the user may make last
minute substitutions or modifications while viewing both the room and the
product specification data as well as pricing.
[0087] To implement the functionality expressed in the foregoing
features and others, a preferred embodiment of the invention enables use
of a knowledge management system to resolve engineering attributes of a
product into visual attributes that the user is able to view. Thus, given a
certain engineering specification, the preferred embodiment is able to
obtain a visual representation of those attributes.
[0088] Further, to facilitate custom room design based on the user's
understanding of how the room will look, a preferred embodiment of the
invention is able to utilize visual attributes of an engineered product to
CA 02588964 2007-05-18
WO 2006/054992 PCT/US2004/038621
identify the closest graphics-renderable component that will represent the
attributes. In doing so, this facilitates accurate custom room design in light
of available product features and constraints.
EXAMPLE
[0089] The following is an exemplary commercial implementation of a
preferred embodiment of the invention. In this implementation,
component attributes are evaluated and an explanation of the manner in
which this assists in customization is provided.
[0090] A number of manufacturers specialize in plumbing-related
fixtures that may be used in a bathroom. One of these manufacturers, for
example, is Kohler. Kohler makes numerous product models, and each
model may differ in size, shape, color, price, material, or other attributes.
A Kohler bathtub, for example, may be 6 feet long, have a whirlpool
feature, have a left-hand drain, be the color white, and cost from $400 to
$1000. This corresponds to a description and the corresponding graphic
renderin.g given in Fig. 12 (bill of materials). With the user having selected
a six-foot bath with a left-hand drain, and located it in a given room layout,
the frame-based functionality or rules-based functionality or both can be
used to make numerous inferences.
[0091] Obviously, a commode cannot now be located in the spatial area
where the tub is located, nor could any other component be located within
the dimensions described by the polygon component that represents the
tub. Further, an inference can be made that for practical reasons a sink
31
CA 02588964 2007-05-18
WO 2006/054992 PCT/US2004/038621
should not be placed directly next to the tub at a certain spatial
orientation,
for this would interfere with both use of the sink and the commode. In
other words, the user should not be required to stand in the tub to use the
sink. This, too, can be accomplished by the inference engine. Besides space
and practical constraints, the inference features may be aware of
engineering constraints related to a particular component and whether and
how it could be used in a particular configuration.
[0092] Attribute constraints are not to be viewed solely as limitations.
Perhaps the user is partial to the Kohler brand. If so, the user might wish
to have as many of the fixtures as possible be Kohler fixtures. The
inference features could be programmed to enable a"Kohler as first
choice" preference. Additionally, the inference features could be
programmed to enable selecting sets of attributes. For example, the user
could elect a preference for Kohler products, and further set a capped price
at a predetermined amount for the objects in the room. In short, the
inference features can be programmed to make multivariable
determinations based on user input.
[0093] The inference features can be object attribute-based, rule-based or
both. Further, the attribute features and rules features are adapted to
interact with a database. The data populating the database can be derived
from multiple sources. For example, database (e.g., database 108 of Fig. 1)
can contain data from vendors, such as product specifications like size or
other attributes. Further, database 108 can also contain data generated
from the customer, such as the size of the room to be configured or color
preferences. Still further, database 108 can contain data obtained from a
32
CA 02588964 2007-05-18
WO 2006/054992 PCT/US2004/038621
retailer, e.g., which brands are most up-to-date, most reliable, provide the
best value, etc. Accordingly, the database can be an extensive library of
accumulated product information from diverse sources, including the
vendor, consumer, and retailer.
[0094] Photographs of particular products can be embedded in the
database 108. Because the visual rendering of the room can be based on
photographs of actual products, greater photorealistic room views are
possible enabling optimal user customization.
[0095] An exemplary sequence of options presented to the user of the
exemplary commercial implementation in configuring a room is illustrated
in Fig. 15. A user can begin at a start page 1500. From here, the user can
choose a room type 1502, e.g., bathroom, and room style 1506, e.g.,
Gentleman's Retreat. This is one method of selecting room attributes 302.
Or, from start page 1500 a user may enter a room ID 1504. This is one
method of configuring a room based on an advertisement or in-store
display 304. Optionally, a user may build a custom room 1508 (i.e., draw
one's own layout or import a CAD file). This is one method of building a
room from "scratch" 306.
[0096] After choosing a room style 1506, a user can choose a room layout
1510, e.g., L-shaped, square, etc. Upon choosing a room layout 1510, the
user can engage in room configuration 1514, 1516 (e.g., a 2-D photorealistic
picture and/or 3-D view). After engaging in room configuration 1514, 1516,
the user may optionally print the room 1518.
33
CA 02588964 2007-05-18
WO 2006/054992 PCT/US2004/038621
[0097] After entering at start page 1500, the user can enter a room ID
1504. This is useful where the reader wants a completely preconfigured
room. The user may then view and configure the room 1514, 1516.
[0098] Still further, the user may enter at the start page 1500, then build a
room from scratch 1508. The user will then choose components 1512,
which can be added to a shopping cart 1528 where items can be fully
configured in advance of purchase.
[0099] The user may have performed room configuration at an earlier
time. If so, from start page 1500 the user can search for a room or shopping
cart 1522 that has been previously stored. Once the room or shopping cart
has been opened 1524, the user can engage in room configuration 1514, or
configure items by means of a shopping cart 1528.
[0100] Still further, after entering at start page 1500 the user may create a
new shopping cart 1520, and then configure items by means of a shopping
cart 1528 in advance of purchase.
[0101] The shopping cart 1528 can be saved 1526 into storage, including
but not limited to storage in database 108. Or, in configuring a shopping
cart 1528 the user may print a quote 1532 or initiate an ordering process via
shopping cart 1534. Upon initiating an ordering process via shopping cart
1534, the user can be led to a new start page 1536, which may be, but is not
limited to, start page 1500.
[0102] If the project/cart/room is saved 1538, a customer search 1540 may
ensue thait determines if the customer is new or not. If the customer is new
34
CA 02588964 2007-05-18
WO 2006/054992 PCT/US2004/038621
1542, the project/cart/room can then be saved and associated with the
customer.
[0103] In accordance with a preferred embodiment of the invention, one
or more processor-based (or other processing device-based) systems may
be used to implement the modules described or apparerit from the
description herein and to perform the functionality described (or inherent)
herein. For each such system, one or more processors (e.g., central
processing unit (CPU)) are provided for execution of one or more computer
programs stored on any (one or more) known recording mediums. The
processor(s) perform, control, or at least inform the various processing
steps performed by the system in sending and retrieving data to and from
at least one user interface and/or network. A user interface may be
connected directly to a bus or remotely connected through a network (e.g.,
Internet). The network represents (wired or wireless) connection of two or
more devices, whether directly or indirectly connected (e.g., directly
coupling through cable, indirect coupling through one or more hubs or
servers, whether the network is local to the processor-based system,
geographically remote from system, or a distributed combination of
local/remote network components).
[0104] Preferably, one or more of the modules are coupled (directly or
indirectly) to one or more database structures for use in supplying storage
functionality for the modules in accordance with the operations described
(or inherent) herein. The database structures can take any form from an
individual floppy disk drive, hard disk drive, CD-ROM, redundant array
of independent devices (RAID) system, to a network of the same or other
CA 02588964 2007-05-18
WO 2006/054992 PCT/US2004/038621
storage devices. As is well known in the art, the database structures may
be physically connected within the same location, or have one or more
structures remotely located in different locations. Each module may have
dedicated or shared access to one or more database structures locally or
remotely located from the module.
[0105] While preferred embodiments of the invention have been
described and illustrated, it should be apparent that many modifications to
the embodiments and implementations of the invention can be made
without departing from the spirit or scope of the invention.
[0106] It should be readily apparent that any known communication
systems or (wired/wireless) networks (e.g., Internet, intranets, private
bulletin boards, individual local or wide area networks, proprietary chat
rooms, ICQ, IRC channels, instant messaging systems, etc.) using real-time
or non-real-time systems may be utilized.
[0107] The modules described herein, particularly those illustrated or
inherent in the instant disclosure, may be one or more hardware, software,
or hybrid components residing in (or distributed among) one or more local
or remote computer systems. Although the modules are shown or
described as physically separated components, it should be readily
apparent that the modules may be combined or further separated into a
variety of different components, sharing different resources (including
processing units, memory, clock devices, software routines, etc.) as ,
required for the particular implementation of the embodiments disclosed
herein. Indeed, even a single general purpose computer executing a
computer program stored on an article of manufacture (e.g., recording
36
CA 02588964 2007-05-18
WO 2006/054992 PCT/US2004/038621
medium) to produce the functionality and any other memory devices
referred to herein may be utilized to implement the illustrated
embodiments. User interface devices may be any device used to input
and/or output information. The user interface device may be implemented
as a graphical user interface (GUI) containing a display or the like, or may
be a link to other user input/output devices known in the art. Discrete
functionality of the system may be separated (logically or physically) to
more efficiently operate the system.
[0108] In addition, memory units described herein may be any one or
more (integrated or distributed) known storage devices (e.g., Random
Access Memory (RAM), Read Only Memory (ROM), hard disk drive
(HDD), floppy drive, zip drive, compact disk-ROM, DVD, bubble memory,
Redundant Array of Independent Disks (RAID), Network Attached Storage
(NAS), Storage Area Network (SAN), etc.), and may also be one or more
memory devices embedded within a processor, or shared with one or more
of the other components. The computer programs or algorithms described
herein may easily be configured as one or more hardware modules, and the
hardware modules shown may easily be configured as one or more
software modules without departing from the invention. Accordingly, the
invention is not limited by the foregoing description, drawings, or specific
examples enumerated herein.
37
