Note: Descriptions are shown in the official language in which they were submitted.
CA 02816469 2013-05-23
FAULTS AND PERFORMANCE ISSUE PREDICTION
PRIORITY
[0001] This application claims priority to U.S. Patent Application No.
13/755,718 filed
January 31, 2013.
TECHNICAL FIELD
[0002] The disclosure relates to systems and methods for predicting
failures or performance
issues that may be configured for gaming machines and systems.
BACKGROUND
[0003] Maintenance operations on gaming devices are typically reactive,
where repair
occurs after a fault or problem has arisen. Typically, a gaming device is down
and remains that
way until the repair is accomplished. A problem is encountered, usually, by a
player who informs
a human operator at the retailer, e.g., a casino or lottery ticket agent. The
operator may call a
service provider, and turn off or take the gaming device out of service until
a technician arrives
to service or repair the gaming device.
SUMMARY
[0004] In one embodiment, new systems and methods for predictive diagnostic
and
prognostic maintenance systems are provided. In certain embodiments, the
systems and
methods may be configured for use with networked gaming machines. In some
embodiments,
the systems and methods may operate in real time and may detect and analyze
data
representing various indicators of machine performance or a current or future
decrease in
machine performance. The data may represent or be used to identify necessary
or
recommended repair, maintenance or other performance issues.
[0005] In another embodiment, systems and methods are disclosed for
automated analysis
of data regarding machine operation and generation of rules related to
predicting the future
performance, repair and maintenance needs of machines. The systems and methods
may
automatically update existing rules related to machine performance, repair and
maintenance
predictions, or needs.
[0006] In one embodiment, sensors on the machines and related software
monitor,
accumulate, store and share information for efficient maintenance and repair
of gaming
DOCSTOR: 2708465\1
CA 02816469 2013-05-23
systems. Diagnostic and prognostic tests may be run and the results
prioritized to identify and
store data related to predictions or expectations of future problems or
requirement for machine
maintenance or repair. Individual machines may send event codes and sensed
data
corresponding to machine status, problematic issues, machine or system errors
or other
indicators related to the machines or their operation. The sensed data may be
collected and
stored in a database. The system may analyze the event codes and sensed data
received
during a time-window from a particular machine and diagnose the machine. The
system may
also examine the received data to provide prognosis of the machine's status
over a defined
time-horizon. In this way, the system may provide a prediction of future
machine performance
as well as future needs for machine repair, maintenance or upgrades.
[0007] In one embodiment, the system can be dynamic and can adapt to changes
in the
expectations for future machine performance and future needs for machine or
software repair,
maintenance or upgrades, for example based on analysis of machine or software
performance
data.
[0008] In another embodiment, methods and systems are provided to analyze
historical or
ongoing (real time) machine performance data (for example, from a collection
of machines) and
determine a set of rules that may be used in an analysis of machine data to
predict future
machine performance and future needs for machine repair, maintenance or
upgrades.
[0009] The system and methods may provide feedback or self reporting including
data of
machine operation and performance subsequent to machine repair, maintenance or
upgrades
and the data may collectively be used to enhance and adapt the rules used in
analysis of
machine data to predict future machine performance and future needs for
machine repair,
maintenance or upgrades. In some examples, the systems and methods can
increase the
availability of gaming machines, increases the mean time between machine
failures and/or
decreases the examples of machine down time.
[0010] In another embodiment, systems and methods are provided by fully
integrating each
game device with sensors that monitor and record game device problems or
events while
accumulating and logging other diagnostic and prognostic data. Events can be
prioritized, and
business decisions can determine actions based on the priorities. In some
examples, diagnostic
and prognostic tests may be administered and events and trends accumulated and
may be
stored locally and later at a central controller.
[0011] In one embodiment, systems and methods are provided for geographical
mapping
that may be displayed on a GUI that assists service technicians. For example,
thousands of
DOCSTOR: 2708465\1 2
game devices distributed within thousands of retail locations over a wide
geographical area
which may be displayed in this manner.
[0012] In one embodiment systems and methods are provided that include
states
indicating when the device is up-and-running, or when the device is down and
awaiting
repair. In some examples, a third state can indicate that a device has a
"minor" problem that
is predicted to lead to a down state for a gaming device, for example, if left
unaddressed.
The minor problem may be some intermittent operation or a parameter, for
example a
temperature increase that is not found in typical or otherwise properly
operational game
devices. The intermittent operation may be due to operator or game player
abuse or neglect,
unusual environments, like high humidity or dust, normal wear or other
malfunction or
breakage. In one embodiment, an "amber" state along with the "red" and "green"
states may
be provided. These colour states are representative examples, and any other
suitable
designation may be used. The visualization can, in some examples, immediately
informs the
nearest technician, for example via geographical mapping so that the
technician can be fully
prepared for servicing one or several sites in the immediate area. Disclosed
methods and
systems can, in some examples, allow the nearest technician to be alerted, the
site and
nearby sites and maintenance status of each may be displayed, accurate
locations and
traveling directions may be displayed, and the methods may include the step of
assigning
priorities by location, type of faults or problems and maintenance status (for
example the risk
of an actual device shutdown or complete malfunction), as well as financial
aspects of the
sites or machines.
[0013] In one embodiment, systems and methods are provided that will
identify and
record examples or symptoms that indicate a potential fault or failure. A
predetermined time
frame for servicing the device may be provided and a timely service visit may
be scheduled.
[0013a] Accordingly, there is described a method for generating fault
prediction rules, the
method comprising: operating at least one gaming device; receiving, at a
processor that is
not within the at least one gaming device, event data of the at least one
gaming device or
from sensors associated with the at least one gaming device; analyzing, at the
processor,
event data from the at least one gaming device; and generating, at the
processor, at least
one fault prediction rule from the event data based on patterns of gaming
device operation
and faults.
3
CA 2816469 2018-01-03
[0014] The event data may include device malfunctions, device repair or
service, or
device operating parameters, such as temperature, performance test results,
online
accessibility; network availability (wireless or hard wired), sound operation,
display operation,
input device operation (for example buttons or other controllers, or other
software or
hardware operational parameters.
[0015] As further described below, some examples method may also include steps
to modify
a set of one or more preexisting fault prediction rules based on the recently
generated fault
prediction rules or data analysis to form a new or updated fault prediction
rule set.
[0016] In certain embodiments, the methods may include steps involving
triggering
diagnostic tests on a gaming device, for example, in response to the event
data analysis.
Additionally, or alternatively, the method may involve predicting a likelihood
of a device
malfunction and then providing an alert to a gaming server.
[0017] In certain embodiments, the method may include the step of analyzing
device
event data over a time window, which time window may be predetermined, against
the fault
prediction rule set and predicting the likelihood of device malfunction or
performance issue.
The method may also include the step of receiving an indicator of a time
window for the
analysis, for example from a game operator or central controller. In one
embodiment, the
time window is separately determined, for example based on the analysis of
aggregate or
individual device event data.
[0018] In certain embodiments, the method may also include the steps of
continuing to
operate the one or more gaming devices during a period of continued operation
and
analyzing event data from the gaming devices during the period of continued
operation. The
method may also involve generating an additional set of fault prediction rules
event data
analysis with a computer processor based on patterns of gaming device
operation and faults
during the period of continued operation or based on a combination of device
operation and
faults during the period of continued operation and a time period preceding
the period of
continued operation. This method may then include the step of modifying the
preexisting fault
prediction rule set based on the additional set of prediction rules and
forming a modified fault
prediction rule set.
[0019] A method is also described for predicting a performance issue with a
gaming
machine. This method may include the steps of monitoring a plurality of data
elements
representing operational characteristics of a gaming machine and then
analyzing the plurality
4
CA 2816469 2018-01-03
of data elements to determine whether the plurality of data elements are
indicative of a risk of
a performance issue with the gaming machine based on a set of rules for
determining the risk
of a performance issue. Subsequently a determination of whether or not the
data elements
are indicative of a risk of performance issue can be performed, and responsive
to a
determination that one or more of the data elements are indicative of a risk
of a performance
issue with the gaming machine, a signal can be transmitted to indicate the
risk of a
performance issue. The risk of a performance issue may be analyzed for a
particular time
period. The method may include receiving an indication that the gaming machine
has been
serviced and recording servicing data related to the servicing of the gaming
machine. This
servicing data may be included in the plurality of data elements analyzed to
determine
whether the plurality of data elements are indicative of a risk of a
performance issue with the
gaming machine based on a set of rules for determining the risk of a
performance issue.
[0020] In some embodiments, the method may include the steps of determining
whether a
gaming machine has been serviced, for example based on event data related to
the gaming
machine and upon determining that the machine has been serviced, suppressing a
signal
indicating the risk of a performance issue.
[0021] In some embodiments, analyzing the plurality of data elements may
include
filtering the data elements.
[0022] In accordance with certain embodiments, the methods may include
outputting a
communication for receipt by a field technician, a game machine operator or a
malfunction
ticket handling system, where the communication indicates a risk of a
performance issue.
Alternatively, or additionally, the methods may include a server assembling
and
communicating data elements representing risks of performance issues.
[0023] In another embodiment a system is provided to perform any of the
methods
described herein. For example, a system may be provided to monitor and
facilitate an
analysis of data collected from a gaming machine that can be used to determine
a level of
risk of a performance issue with the machine. The system may include a
processor
configured to collect data representing operational aspects of the gaming
machine and
transmit the collected data; a database configured to receive and store the
collected data,
where the database is operatively linked to the gaming machine; and a server
operatively
linked to the database, where the server can be configured to analyze the
collected data
and to generate a risk factor for a performance issue with the gaming machine,
the risk
CA 2816469 2018-01-03
factor being based on an analysis of the collected data and a set of rules for
determining
the risk of a performance issue with the gaming machine. The server may also
be
configured to revise the set of rules for determining the risk of a
performance issue with the
gaming machine based on analysis of gaming machine performance data.
[0024] In another embodiment, a non-transitory computer readable medium
having
instructions stored therein thereon is provided. When executed, the
instructions are operable
to monitor and facilitate an analysis of data collected from a gaming machine
and to
determine a level of risk of a performance issue with the machine and to
generate and a set
of rules for determining the risk of a performance issue with the gaming
machine based on
analysis of gaming machine performance data.
[0025] In another aspect, there is described a system for generating fault
prediction
rules based on an analysis of event data collected from a gaming machine, the
system
comprising: a processor that is configured to collect data representing
operational
aspects of the gaming machine; a computer readable medium that stores a
database that
is configured to receive and store the collected data, the database being
operatively linked to
the gaming machine; a server that is operatively linked to the database and
that is configured
to process the collected data and to generate a risk factor for a performance
issue with the
gaming machine, the risk factor being based on a set of fault prediction rules
for determining
the risk of a performance issue with the gaming machine, wherein the set of
fault prediction
rules is generated by the processor from the event data based on patterns of
gaming
machine operation and faults.
[0026] In accordance with another aspect of the present disclosure, there
is provided a
method of predicting a performance issue with a gaming machine. The method
includes
receiving, at a processor, a plurality of data elements representing
operational
characteristics of the gaming machine; and upon determining that the plurality
of data
elements are indicative of a risk of the performance issue with the gaming
machine based
on a set of rules for determining the risk of the performance issue,
generating an alert
indicating the risk of the performance issue.
[0027] In accordance with another aspect of the present disclosure, there
is provided a
system for determining a level of risk of a performance issue with a gaming
machine. The
system includes a database configured to store data representing operational
aspects of the
gaming machine; and a server configured to generate a risk factor for a
performance issue
6
CA 2816469 2018-12-06
with the gaming machine, the risk factor being based on the stored data and a
set of rules for
determining the risk of a performance issue with the gaming machine.
BRIEF DESCRIPTION OF THE FIGURES
[0028] Example embodiments of the invention are illustrated in the
accompanying
drawings in which:
[0029] Figure 1 is a schematic, block diagram of a network with client game
devices, a
retailer, and a central controller;
[0030] Figure 2 is a schematic, block diagram of a typical game device,-
retailer, and
central controller;
[0031] Figure 3 is a flow chart of operations according one embodiment of
the invention;
[0032] Figure 4 is a timeline showing the relationship of certain events
according to
certain embodiments;
[0033] Figure 5 is another timeline showing the relationship of certain
events according to
certain embodiments;
[0034] Figure 6 depicts a block diagram showing certain components of an
exemplary
gaming machine.
6a
CA 2816469 2018-12-06
CA 02816469 2016-01-08
[0035] Figure 7 depicts a flowchart illustrating an exemplary method of
facilitating machine
performance predictions.
[0036] Figure 8 depicts a flowchart illustrating another exemplary method
of facilitating
machine performance predictions.
[0037] Figure 9 depicts a flowchart illustrating another exemplary method
of facilitating
machine performance predictions.
DETAILED DESCRIPTION
[0038] For simplicity and illustrative purposes, the principles of the
present invention are
described by referring mainly to various exemplary embodiments thereof.
Although certain
example embodiments of the invention are particularly disclosed herein, one of
ordinary skill in
the art will readily recognize that the same principles are equally applicable
to, and can be
implemented in other systems. Throughout this description, certain acronyms
and shorthand
notations are used. These acronyms and shorthand notations are intended to
assist in
communicating the ideas expressed herein and are not intended to limit the
scope of the
present invention. Other terminology used herein is for the purpose of
description and not of
limitation.
[0039] Figure 1 is a schematic block diagram of a wide area computer
network 2 that may
include a plurality of game devices 4 communicating with a retailer 6 over a
local area network 5
(LAN). The retailer communicates via the Internet 8 or other WAN to a central
controller 10. A
technician 12 may be contacted via the same communications network.
[0040] Game devices 4, retailers 6 and a central controller 10 may be in
the form of
computer servers or processors that may run one or more applications on one or
more
platforms (hardware and software) suitable for gaming applications.
[0041] In some examples, game devices 4 can be VLTs (video lottery
terminals) connected
to retailers 6 (a retailer is a terminal that may act as a site or LAN
controller) via LAN 5.
[0042] In some examples, game devices 4 ca be networked with central
controllers 10
directly or via a peer to peer network.
[0043] In some examples, a central controller can be located at and/or
operated by a retailer.
In some examples, the central controller can be part of a game device.
7
CA 02816469 2013-05-23
[0044] In some examples, the technician 12 can be located at the central
controller 10. In
some examples the central controller 10 and be operated by the technician 12.
[0045] Figure 2 is an illustrative block diagram of hardware and certain
software modules
that may be found in the game device 4, retailer 6, and/or central controller
of Figure 2. The
central controller may have a more powerful implementation that performs many
operations at
high speeds, while the retailer may have more modest in hardware and
processing speed, while
the game device may have relatively modest hardware and software. Regardless,
similar
hardware may be found in each. In some examples, a computer bus 20 connects a
processor
22 to memory 24; to I/O (Input/Output) interfaces 26; and to communications
hardware 28 that
connects to the network 30 (here the NETWORK 30 may be the Internet 8 or the
LAN).
[0046] The processor 22 may be any processor or controller or control logic
arranged to
execute program code and exercise control over the retailer, central
controller or game device.
Processors made by Intel, AMD, or any other manufacturer may be used, as well
as ASIC.
(Application Specific Integrated Circuit) or other particular designs. In some
examples, the
memory 24 can includes a ROM and/or a RAM. Again, standard hardware may be
used,
including electronic or magnetic ROM and RAM, flash, optical, CD's, hard disk
drives, etc.
External memory, not shown, may be used and include disk and RAID systems. The
I/O
interfaces 28 may include drives for motors. LED and other displays, printers,
touch screens,
mouse and keyboard or key, and other such interfaces.
[0047] The communications hardware 28 may include drivers for discrete wires,
twisted
pairs, Ethernets, Optical fiber, wireless and any other transmission types
known to those skilled
in the art.
[0048] In some examples, a video display and/or a touch screen keyboard 25
can provide
visual information for playing the game in a game device, or controlling and
monitoring the
devices on an LAN for the retailer. Such a video display may operate as a
graphical users
interface (GUI).
[0049] In Figure 2, the example memory is shown having: a data storage 130
for local
operations, a communications program stack 132 that implements the
communications
hierarchy; an operating system 134, device drivers 136, memory and other
system managers
138 and local game control and operation software 140. Other configurations
can also be used.
[0050] Figure 3 is an example flow chart showing blocks associated with a
method for
generating fault prediction rules. In some examples, the method can be
performed by a game
device, a retailer, a central controller or a combination thereof. A game
device, such as VLT 40,
can detects and logs data associated with events that include issues or
problems associated
DOCSTOR: 2708465\1 8
CA 02816469 2013-05-23
with the game device. These can include, for example, access doors, bill
(currency) acceptors,
printers, coin mechanisms, button or touch sensors, the device game itself,
power supplies,
software and hardware errors, and the like. In some examples, the game device
can include
one or more sensors for detecting issues or problems, and can, in some
examples, generate
event data. In some examples, components of the game device can generate
signals,
interrupts, or messages indicating issues or problems. In some examples, the
data related to
the events may include results from diagnostic and prognostic tests run on the
game devices.
[0051] For example, in Figure 3, the data acquisition 42 can collects data
on the video
monitor 44, the printer 46 and the bill acceptor 48. The data from these
input/output devices
may be logged, tracked, processed 50 and/or tagged with a priority. A
condition monitor 52 can
apply prior established thresholds (developed heuristically or with historical
data, for example)
and can compare them to the data. Self generated diagnostics 56 or prognostic
58 can be
initiated by the retailer, the central controller, the VLT itself or a
technician, and can test
functions and operations of the VLT. In some examples results can be tracked
and stored.
[0052] Prognostic tests 58 may be run to identify parts of the VLT that are
at risk of future
failure. Prognostics tests may be scheduled to run at recurring times, for
example on a
predetermined schedule or they may be performed on demand as required. In some
examples,
the data from these tests can generate decisions 60 on how to proceed. The
responses 64 to
these decisions may include communication among or between the retailer, the
VLT system
itself, the central controller, and/or a technician. In some examples, the
graphical user interface
(GUI) 62 can provide a display where the data, problem and/or decisions 60 for
review by a
technician or other personnel. In some examples, In some examples, the GUI can
provide feed
back or controls options for an operator to adjust reactions or responses of
the system.
[0053] In some examples, event data can be specific to a particular gaming
site, such as a
casino or a geographic location, or to a particular gaming machine, device, or
model of gaming
device. In one embodiment, the event data represents at least one malfunction
selected from
the group consisting of mechanical, electrical and software malfunctions or
errors. The event
data may also represent machine status, certain problematic issues or machine
errors.
[0054] In some examples, data analysis, such as trending or clustering, can
be used for
prognosis determinations. For example, diagnostic and prognostic tests may be
self
administered locally within the LAN or by the VLTs themselves. In some
examples, low priority
events and trend data can be maintained in the LAN and sent to a central
controller when some
higher priority event is sent, or when the network traffic is low.
DOCSTOR: 2708465\1 9
CA 02816469 2013-05-23
.=
[0055] In some examples, low priority and/or trend events can include,
inter alia: higher than
expected temperatures, printer paper getting low or jamming, a bill acceptor
rejecting too many
bills, button sticking, intermittent problems closing doors, or glitches on a
display screen, etc.
Low priority events can include events that do not disable the operation of a
game device, but
can be repaired at a later time, for example, because they negatively affect
the operation of the
game device or the game playing experience.
[0056] In some examples, high priority events suggest or call for more
immediate attention
and/or repair. Such events can include any event that shuts down the game
device, e.g., bill
acceptor not working, a video display not working, the VLT not working, etc.
These events can
be reported via the retailer to the central controller immediately. Some
example methods
disclosed herein can include escalating the reporting of a predicted fault to
game operators, field
technicians or other personnel. In some examples, ticketing systems can
operate to see that the
game machine is serviced or repaired as quickly as possible.
[0057] In some examples, the central controller can coordinate and respond to
reported
problems in a comprehensive manner. For example, a high priority problem
arises and is
reported to the central controller, via the retailer, which may send the lower
priority, trend and
diagnostic data along with the high priority event. The central controller may
send low priority
data regarding all the game devices at a site. If a service technician visit
is necessary, the
central controller may automatically inform the service technician about high
and all the low
priority issues at that site. The central controller may also automatically
inform the service
technician about low priority issues at other sites that will be on his
geographic route to reach
the site with the high priority problem.
[0058] In some examples, before a service technician visits a retailer
location, the system
can provide a technician with specific problems and/or symptoms related to all
of the game
devices at that site. The technician can arrive with all the tools, including
diagnostic software,
and replacement parts specific to the problems to be addressed. In some
examples, this can
improve maintenance system efficiency. In some examples, a GUI may be
available for a
technician to enter test/diagnostic/prognostic programs.
[0059] In some examples, a set of business rules 54 can be configured for
facilitating
responses to the data results for efficient maintenance of the gaming systems.
In some
examples, business rules may be applied that enhance the decision for types of
responses to
the problems. These rules may be amended to accommodate unforeseen problem
that may
arise.
DOCSTOR: 2708465\1 10
CA 02816469 2013-05-23
[0060] In some examples, low priority and informational data can be buffered
(stored) at the
local game device or the retailer and sent to the central controller during
low network traffic
times. Informational data may include accounting information that is usually
sent at the end of
the business day. In some examples, this can preserve bandwidth or can avoid
overloading
systems at peak traffic times. In some examples, this can be applied to low
bandwidth networks
such as dial-up, cellular, satellite, leased lines or wireless, or to networks
wherein costs vary
based on the time of day.
[0061] A partial list of device events incorporated into an illustrative
example of the present
invention includes, for example: any access door open; the bill validator is:
empty, miscounted,
jammed, cheated, timed out, full, etc.; the printer is: offline, paper is low,
no communications,
paper jammed, cutter jammed, mechanism opened, misaligned, timed out (meaning
some
function had a time limit that expired), etc.; the coin mechanism is: mal-
functioned, jammed,
empty, electronic signal in error, etc.; touch sensors not responding, cash
out buffer full;
terminal disabled by retailer or central controller; event buffer full;
batteries are low, games not
installed or enabled or operating; a general system error; the system
experienced a power down
event; outdated versions of software or device drivers. When diagnostics are
exercised other
conditions that may be buffered include cooling fan speed, internal
temperatures, presence and
use of Flash memory, etc.
[0062] In Figure 3 business rules 54 may include the following: a game device
and the
retailer buffer events including low and informational events and once a day
(or weekly or
biweekly or monthly or twice a day), the retailer will send the buffered data
to the central
controller at low network traffic time; whereas high priority events may be
sent by the retailer
immediately to the central controller; an authorized system person may be able
to initiate and
view the results of a self test or diagnostic for a game device from the
central controller, the
retailer or the game device itself; the central controller may be able to
interrogate the retailer or
game device for bill acceptor status, memory status, temperature, battery
status; the data from
self tests and diagnostics may be buffered for reporting and trending
analyses. Trending data
may include identifying and tracking usage of peripherals, electronic
components, sensors, etc.
In some examples, virtually all components found in a game device or retailer
and the estimated
or mean time between failures may be determined and displayed.
[0063] Priority can be shared by the game device, the retailer and the
central controller.
Priority can be high when, for example, business rules so determine. For
example, high priority
can be assigned when a game device is not operating or is malfunctioning in a
way that
prevents wagering activity or game playing or the machine is otherwise
prevented from
DOCSTOR: 27O84651 11
CA 02816469 2013-05-23
=
generating income, or the same or a similar situation occurs for an entire
site of several game
devices, or an entire LAN, an entire site or an entire gaming system. For
example, in addition to
the sensors on a game machine one or more sensors may be placed on the network
to confirm
that the game machines are operating, for example by a game machine returning
an
appropriate response following receipt of a status check message at the game
machine. The
system can also be configured to allow for prioritization so that some sites,
some LANs, or some
game devices are assigned higher priorities. For example, those with greater
commercial value
may be assigned higher priority. Moreover, some geographic location may be
more visible to
the public than others or may be more important, for some reason, than others
and so may be
assigned higher priorities. The human owners of a gaming system or site may
set their priorities
as they see fit.
[0064] The business rules 54 may include buffered data mining where programs
may
process the buffered data to determine and identify key attributes of hardware
and/or software
failures that presage future failures. Predictive events may be visualized and
maintenance steps
taken or modified to minimize the effects of future failures. In this aspect,
the business rules 54
may be amended and/or added to so as to efficiently implement maintenance
procedures that
minimize these future effects.
[0065] Accumulating data may include measuring and storing many different
operating
parameters and characteristics of game devices and LANs, etc. Trends in these
parameters and
characteristics may suggest preventative maintenance that pre-empts more
costly future failure
that might have occurred. Measuring and tracking power usage, down time,
temperatures, time
that a game device may be being played, or upcoming special usages or
occasions may dictate
that preventative maintenance. One other important parameter may be the amount
of time and
type of servicing a technician may have spent on a game device, or at a site
or on an LAN or
WAN. These and other similar data allow for analysis that provides predictions
for service and
repair, leading to an effective, efficient maintenance system.
[0066] In another embodiment, an Online Adaptive Diagnostics and Prognostics
(OADP)
system and methods are disclosed. In some embodiments, the OADP system and
methods can
be designed for gaming machines. The individual gaming machines can send event
codes and
sensed data corresponding to status and/or problematic issues on the machines.
The event
codes and sensed data may be stored in a database. The disclosed system
examines the event
codes and sensed data received during a time-window (for example a
predetermined or
otherwise externally determined time frame) from a particular machine and
diagnoses the
DOCSTOR: 2708465\1 12
CA 02816469 2013-05-23
machine. The system also examines the received data to predict the machine
status over a time
horizon or at a defined point in the future.
[0067] In some examples, the failure prediction systems and methods described
herein can
improve system uptime and availability by predicting the occurrence of certain
types of
hardware and software errors, thereby increasing system and gaming machine
mean time
between failures. In some examples, some machine faults can be predicted
before they occur
by comparing machine events and sensor data with a set of rules. The set of
rules may be
dynamic and generated by an off-line process which examines the historical
event codes and
related data produced by different electronic gaming machines.
[0068] In some examples, a feedback system or mechanism (which may consist
entirely or
partly of software) can be used to enhance and adapt the generated set of
rules. Accordingly,
the set of rules may be dynamic and the system and methods increase the
availability of gaming
machines and increase the mean time between failures or other down time for
the gaming
machines. Overall repair and maintenance costs may also be reduced since
certain costly
repairs may be avoided.
[0069] Figure 4 is an example timeline 100 showing the relationship of
certain events in
accordance with example methods and systems of the application. In the figure,
horizontal line
102 shows increasing time. In this embodiment, a method of failure prediction
may be
summarized as follows. Given the current state of an electronic gaming machine
at time t 104,
the method may predict the potential occurrence of a failure at a future time,
if 106, as shown in
Figure 4. The current state of a machine at present time t 104, can be
represented by the set of
events that occurred within a data window (time-window) preceding t 104.
Potential occurrence
of a failure within a prediction time window (time-horizon) can be predicted
based on the event
patterns and data that occur within the time window and may also include event
patterns and
data that precede the time window. Data mining techniques may be applied to
identify and
recognize patterns in the event data within the time window by analyzing a
large set of event
data. In some examples, the method can determine or identify the events or
data that commonly
leads to machine failures within the time-horizon using data mining
techniques.
[0070] Figure 5 is an example timeline 120 depicting a series of events
122a, 122b and 122e
that occur before a failure f 124, as shown in Figure 5. As with Figure 4, in
Figure 5, horizontal
line 102 shows increasing time. Events 122a, 122b, and 122e occur in varying
frequencies
within a time window 124, leading up to a machine failure f 126 within a time
horizon 128. In one
embodiment, the systems and methods can predict the likelihood or risk of a
failure within the
DOCSTOR: 2708465\1 13
CA 02816469 2013-05-23
time horizon 128. In another embodiment, the systems and methods can predict
the likelihood
or risk of a failure at a particular time point, such as time point t 130.
[0071] An exemplary method to identify the failure patterns includes analyzing
a large or
otherwise statistically significant set of historical event data, referred to
as training data. In one
embodiment, the training data includes a set of event data from electronic
gaming machines
received by a device such as a central controller, though similar methods may
rely on data from
other machines. In some examples, the central controller can performs data
analysis that results
in the generation of a set of rules. The rule generation can be time consuming
due to the large
size of the training data; hence, in some examples, it may run as an off-line
process. In some
examples, generating rules based on a training data set can include
customization for different
deployment cases. For example, the analysis can incorporate factors such as
different types or
models of gaming machines, the length of time that individual machines have
been in the field,
different models of peripherals in the gaming machines, different types of
venues where the
machines are installed and the maintenance or repair history of the different
machines.
[0072] In one embodiment, the size of the training data may be determined by
three factors;
(i) the event data time window 124; (ii) the time horizon 128 for prediction;
and (iii) the employed
protocol. The time window 124 can identify and specify the length and identity
of the time block
the training is to consider and analyze when examining the historical data.
The time horizon 128
can specify the prediction horizon, that is the length and identity of the
time block in the future
that the system is to consider in determining predictions. In some examples,
increasing the time
horizon 128 may enhance the probability that a failure is predicted correctly,
for example, when
a failure is predicted. On the other hand, in some examples, if the time
horizon 128 is too large,
the prediction may be of limited use because it is not clear when the failure
might occur. The
employed protocol can define the event codes to be analyzed. The employed
protocol can
define the features used in training. Different protocols may have a different
number and set of
event codes. For example, one protocol may include as many as 400 different
event codes, or
even more.
[0073] In some embodiments, the set of rules used for analysis can be
generated off-line in
a process that is separate from live or ongoing game play or machine
operation. The data set
used to generate the rules may be fixed and unchanged throughout the data
analysis steps. In
other embodiments, the set of rules can be generated online in an adaptive
technique, for
example, during live game play or where the data set being analyzed is dynamic
and updated,
for example with regular or irregular updates, which may occur on a monthly,
weekly, daily,
hourly or even more frequent basis.
DOCSTOR: 270846511 14
CA 02816469 2013-05-23
= '
[0074] In some examples, generating rules online or with an adaptive process
can require
less time and resource consuming data analysis. In the online or adaptive
process the data can
be automatically collected from gaming machine sites, for example, through the
network. In an
adaptive technique, a set of rules can be already in place and the systems and
methods can
operate to continue to monitor data from the gaming machines and update the
existing rules. As
a result, in some examples, the amount of data analyzed to update a set of
rules may be less
than amount of data necessary to create an initial set of rules from scratch.
[0075] Generating the training data may include the step of preprocessing the
data, where
software in the system may be configured to perform a preprocessing step,
involving cleaning
the collected data with a preprocessor. Cleaning the data may involve any one
or more of the
following subtasks: noise reduction or removal, identification and removal of
outlying data
entries, and resolving inconsistencies in the data. Cleaning may also refer to
taking data in a
raw or uncleaned state or form and converting the data into a form that is
better suited for later
data analysis steps, for example, data mining or modeling tasks. For example,
cleaning may
include processing or removal of extraneous or unnecessary data such as meta
data, tags, or
empty fields. Software in the system may also be configured to filter the
collected data. In this
context, filtering can include feature extraction where redundancies (i.e.,
attributes carrying
duplicate or less information) can be eliminated by a function or ranking
process. Other
techniques for data manipulation may also be used or they may be used in the
alternative, for
example wrapper, embedded and search based models of data management and
manipulation.
The preprocessing step may be performed separately, in sequence or in
parallel, or together
with other steps. Similarly, the software module or engine(s) that perform
these steps may be
provided separately or together, for example filtering the data.
[0076] In an adaptive technique, the time for the preprocessing the data
may be reduced,
since the total amount of data being processed may be reduced, when compared
against the
amount of data that would need to be preprocessed in the absence of a
preexisting set of rules.
[0077] In some examples, adaptive or online techniques may be configured to
accommodate
and responsively adjust to any mismatch arising when installing the prediction
software at a new
site. For example, when first installed at a new site, there may be a mismatch
between the
training data set which was generated using historical data from another site
and the gaming
machines and data being produced at the new site. The adaptive techniques may
be configured
to avoid significant negative impact on the performance of the diagnostics and
failure prediction
or other issue prediction systems disclosed herein.
DOCSTOR: 270846511 15
CA 02816469 2013-05-23
A=
[0078] In some examples, adaptive or online techniques can accommodate new
features
and data codes in the data set being generated by the gaming machines.
Advances in
electronic gaming machines, advances and developments in gaming machine
diagnostics, as
well as the use of new protocols, may all result in new data codes for new
events being
generated. This new data may necessitate generating new rules as well as new
decision trees
to handle and make machine operation performance predictions based on this new
data.
[0079] The adaptive or online techniques can, in some examples, allow game
operators to
selectively include or exclude certain data codes for events from the decision
process. The
adaptive techniques can operate in a flexible way to allow game operators this
level of control
and may even allow the system to generate new prediction rules or update
existing rules based
on a revised set of data codes or selected events.
[0080] The adaptive or online techniques can, in some examples, facilitate
handling
heterogeneous event codes (e.g., event codes from sites which have different
models of
machines or different machines manufactured by different vendors).
[0081] In certain embodiments, the adaptive or online techniques may
eliminate training
steps. For example in the case of a new installation of the performance
prediction software
system, a set of preexisting rules may be used as a starting point, and the
adaptive system can
then customize and update that set of rules to reflect the unique experience
and history of that
gaming site as time passes and gaming machine performance data is generated
and even as
the gaming site evolves with new gaming machines being added and older gaming
machines
being removed, or even changes in the population of different features and
options appearing in
the suite of gaming machines at the site.
[0082] Figure 6 depicts a block diagram showing certain components of an
example gaming
machine 200. The gaming machine 200 may include a printer 210, a bill acceptor
212, a card
reader 214, a display 216, a CPU temperature sensor 218 and a door sensor 220.
A gaming
machine may be provided with various other sensors and components or it may
also be
provided with fewer sensors and components. In addition to the CPU temperature
sensor 218,
various other temperature sensors (not shown) may be provided, for example to
monitor the
temperature of other components of the gaming machine or the temperature of
the gaming
machine generally. The door sensor 220 may be useful to indicate to a central
controller
whether the gaming machine door is closed or open, the latter position
indicating a possible
security breach.
[0083] Figure 7 depicts a flowchart illustrating an example method 300
of facilitating machine
performance predictions in accordance with certain embodiments of the
invention. In particular,
DOCSTOR: 270846511 16
CA 02816469 2013-05-23
Figure 7 shows an example method of predicting potential failures in gaming
machines. The
example method illustrates two processes: a rule generation process and
failure prediction
process. The rules generation process can be used to characterize the failure
and non-failure
patterns of the gaming machines using data from the gaming machines which is
collected in the
form of event codes 310. A software module or collection of software and
hardware
components, such as sensors, can send data to an event codes collector 312.
Further details
related to the source and transmission of event code data to the event codes
collector 312 is
provided below.
[0084] The data representing the event codes may be stored in a database, and
in some
embodiments the event codes 310 may be stored in the database. A rules
generator 314 can be
in communication with the event codes database 310, as applicable, and may
generate an initial
set of rules using the existing event codes. Alternatively, the rules
generator 314 may use a pre-
generated set of rules as an initial online rules set 316. The rules generator
314 can adapt the
online rules set 316, for example, in real time, depending on the observed
event codes and the
feedback information it receives through the event codes collector 312 related
to the failure
prediction.
[0085] The online rules set 316 can contain the set of rules to predict
failures or other events
such as performance issues. Initially this could be a set of predefined set of
rules that has been
generated off-line. The online rules 316 may be represented as a set R={r1,
r2, T3, ..., rm}. In
some examples, the set of rules can be dynamic and the number of elements in
the set of rules
R, IN, may change depending on the collected event codes. The rules generator
314 may
generate the online rules 316. Alternatively, or additionally, the rules
generator 314 may also
update an existing set of online rules 316, for example based on ongoing
analysis of the event
codes database 310. A clustering technique, trend analysis, decision tree, or
any other suitable
data analysis algorithm or collection thereof may be employed for this
purpose. For example, in
one embodiment the rule generator 314 may examine data leading up to a gaming
machine
failure to determine whether there might be identifiable trigger events, such
as repeated
occurrences of one or more event codes 310 or particular combinations of event
codes 310 or
examples of two or more event codes occurring within a particular time frame
or examples of
three or more event codes occurring within a particular time frame.
[0086] The event codes collector 312 can collect event data from the different
gaming
machines 318a, 318b, 318c and store them into the event codes database 310,
for example
upon receiving the events via the network 320. Events data may be used for
predicting a failure
DOCSTOR: 2708465\1 17
CA 02816469 2013-05-23
320 or predicting a different performance related event and may include
firmware fault, optical
fault, component fault, and memory fault, among others.
[0087] In step 322 of the method, a determination can be made 322 as to
whether a fault (or
other performance event) is predicted for a given gaming machine, for example,
within a given
time window or by a particular time point in the future. If the determination
is that no fault (or
other performance event) is predicted 324, then the method can continue to
collect additional
event codes with the event codes collector 312.
[0088] If
step 322 results in a prediction of a fault (or other performance event) 326
then the
method may proceed to perform diagnostics 328 on the gaming machine predicted
to have a
fault.
[0089] In one embodiment, the fault prediction step 332 can operate as
follows: for each
gaming machine, the system can analyze the set of events that occurred during
a defined time
window and compare them against the online fault prediction rules 316. If a
machine is
expected to fail or experience a performance issue during the future time
window, the system
can generate an alert. The alert can be a message sent to technician via e-
mail, SMS, etc. In
some examples, the system can trigger a situation manager 330 to display an
alert. In some
examples, the alert can trigger the issuance of a maintenance ticket in a help
desk system.
[0090] The failure prediction step 322 may include considering repair data
related to
components that have been repaired or serviced such that the system will not
generate false
alarms and request additional unnecessary repairs for such machines. This
repair data may be
maintained in a separate database (not shown) or it may be maintained within
or as part of the
event codes database 310. If a gaming machine has been serviced, repaired, or
even replaced
during the time window the failure prediction may be configured to suppress
any alert
generation for that particular machine.
[0091] The fault prediction alerts may be used to trigger diagnostic tests 328
to be performed
by or on a gaming machine to verify failure predictions or performance
predictions. The fault
prediction engine with the adaptive method may also be applied in a root cause
analysis, for
example, in an off-line mode, to analyze and predict faults in the repair shop
of the operator.
[0092] The fault prediction analysis 322 may involve discovering which rule or
rules match
the current state of a gaming machine. The current state of a gaming machine
may be
represented by a collection of events codes and corresponding attributes. The
fault prediction
analysis 322 can determine a match or best fit for known failures and the
closest match for
unknown failures. If R is the set of rules, in the online rules 316, where
R={r1, r1, rm} and
the current state of a gaming machine is Sj={e1t, e2t, ent}
including the set of events and the
DOCSTOR: 2708465 \ 1 18
CA 02816469 2013-05-23
corresponding attributes values collected at time t. A similarity analysis may
be used to find the
best rule that matches the current state Sj.
[0093] The situation manager 330 may provide feedback to the rules generator
314 thereby
allowing a game machine operator to specify false alarms and ignore or prevent
certain
predictions. In some examples, the rules generator 314 can include feedback
information the
next time it generates the online rules 316 and thereby can adapt the rules to
reduce false
alarms and increase prediction accuracy. Over a period of time, in some
examples, multiple
rounds of event code 310 analysis and rule generating the system 300 can
optimize the online
rules 316 for predicting relevant faults or performance issues. In some
examples, this may be
achieved, through normalizing and developing a baseline for normal operation
of a gaming
machine and detecting any deviation towards a fault in the future based on any
monitored event
from the gaming machine.
[0094] Figure 8 depicts a flowchart illustrating an example method 400 for
generating fault
prediction rules. At event 412, event data can be received or collected from
the gaming devices
or from sensors associated with the gaming devices. In some examples, event
data from the
gaming devices can be analyzed 414 with a computer processor to generate 416
one or more
fault prediction rules based on the event data and patterns of gaming device
operation and
faults. The event data may include device malfunctions, device repair or
service, or device
operating parameters, such as temperature, performance test results, online
accessibility;
network availability (wireless or hard wired), sound operation, display
operation, input device
operation (for example buttons or other controllers), or other software or
hardware operational
parameters. In some examples, the fault prediction rules can be optionally
updated 418 and in
some examples, the method may then proceed back to the beginning 420, where
the newly
updated rules may be further updated based on additional data collected by the
system.
[0095] Figure 9 depicts a flowchart illustrating an example method 500 for
predicting a
performance issue with a gaming machine. This method may include receiving 510
data
elements representing operational characteristics of a gaming machine. In some
examples, the
data elements can be analyzed 512 to determine 514 whether the plurality of
data elements are
indicative of a risk of a performance issue with the gaming machine 514. As
described herein
this determination can be based on a set of rules for determining the risk of
a performance
issue. Upon determining that the data elements are indicative of a risk of
performance issue, an
alert can be generated 516 to indicate the risk of a performance issue. In
some examples, the
alert can be a signal or message sent to technician via e-mail, SMS, etc. In
some examples, the
DOCSTOR: 2708465\1 19
CA 02816469 2013-05-23
system can trigger a situation manager 330 to display an alert. In some
examples, the alert can
trigger the issuance of a maintenance ticket in a help desk system.
[0096] In some examples, the risk of a performance issue may be analyzed for a
particular
time period.
[0097] Additionally, in some examples, the method may include receiving a data
element
indicating that the gaming machine has been serviced and recording servicing
data related to
the servicing of the gaming machine. The method may also include filtering any
predicted faults
to remove any false positives, for example if a machine was newly deployed or
had been
recently serviced, or other known false positive result. The methods may also
include prioritizing
the determined performance issues based on a severity of the performance issue
or an
assessed risk of game machine malfunction; as well as escalating the reporting
of the
determined performance issues to game operators, service technicians or
malfunction ticketing
systems.
[0098] The methods may also include the steps of providing a report of the
determined
performance issues via a graphical user interface, for example to a business
intelligence unit for
review by game operator executives, marketing personnel and other game
operations persons
or regulators.
[0099] The disclosed invention is applicable to both hardware and software
failures and is
applicable to gaming devices in online gaming, government sponsored and
commercial gaming
environments.
[00100] The methods described herein can be implemented with gaming devices 4
networked
with a central controller 10, as illustrated for example in Fig. 1. In other
examples, the methods
can be implemented at a single site or retailer 6. The methods can also be
implemented on a
single gaming machine having a processor. In some examples, the methods can be
implemented on multiple gaming machines networked together with one or more of
the gaming
machines having processors. In other examples, the methods can be implemented
on any
combination of these elements or other suitable combinations.
(00101] The above-described embodiments of the present invention can be
implemented in
any of numerous ways. For example, the embodiments may be implemented using
hardware,
software or a suitable combination thereof. When implemented in software, the
software code
can be executed on any suitable processor or collection of processors, whether
provided in a
single computer or distributed among multiple computers. Such processors may
be
DOCSTOR: 270846511 20
CA 02816469 2013-05-23
implemented as integrated circuits, with one or more processors in an
integrated circuit
component. Further, a processor may be implemented using circuitry in any
suitable format.
[00102] It should be appreciated that a computer may be embodied in any of a
number of
forms, such as a rack-mounted computer, a desktop computer, a laptop computer,
or a tablet
computer. The disclosed systems and methods are not restricted to electronic
gaming
machines, they would also apply to any functionally equivalent devices that
can be used for
gaming purposes such as personal computers, laptops, tablets, personal digital
assistants
(PDAs), mobile devices, a Web TV, a smart phone or any other suitable portable
or fixed
electronic device.
[00103] Also, a computer may have one or more input and output devices. These
devices can
be used, among other things, to present a user interface. Examples of output
devices that can
be used to provide a user interface include printers or display screens for
visual presentation of
output and speakers or other sound generating devices for audible presentation
of output.
Examples of input devices that can be used for a user interface include
keyboards, and pointing
devices, such as mice, touch pads, and digitizing tablets. As another example,
a computer may
receive input information through speech recognition or in other audible
format.
[00104] Such computers may be interconnected by one or more networks in any
suitable
form, including as a local area network or a wide area network, such as an
enterprise network or
the Internet. Such networks may be based on any suitable technology and may
operate
according to any suitable protocol and may include wireless networks, wired
networks or fiber
optic networks. As used herein, the term "online" refers to such networked
systems, including
computers networked using, e.g., dedicated lines, telephone lines, cable or
ISDN lines as well
as wireless transmissions. Online systems include remote computers using,
e.g., a local area
network (LAN), a wide area network (WAN), the Internet, as well as various
combinations of the
foregoing. Suitable user devices may connect to a network for example, any
computing device
that is capable of communicating over a network, such as a desktop, laptop or
notebook
computer, a mobile station or terminal, an entertainment appliance, a set-top
box in
communication with a display device, a wireless device such as a phone or
smartphone, a
game console, etc. The term "online gaming" refers to those systems and
methods that make
use of such a network to allow a game player to make use of and engage in
gaming activity
through networked, or online systems, both remote and local. For example,
"online gaming"
includes gaming activity that is made available through a website on the
Internet.
[00105] Also, the various methods or processes outlined herein may be coded as
software
that is executable on one or more processors that employ any one of a variety
of operating
DOCSTOR: 270846511 21
CA 02816469 2013-05-23
,
, . systems or platforms. Additionally, such software may be written
using any of a number of
suitable programming languages and/or programming or scripting tools, and also
may be
compiled as executable machine language code or intermediate code that is
executed on a
framework or virtual machine.
[00106] In this respect, the invention may be embodied as a tangible, non-
transitory computer
readable storage medium (or multiple computer readable storage media) (e.g., a
computer
memory, one or more floppy discs, compact discs (CD), optical discs, digital
video disks (DVD),
magnetic tapes, flash memories, circuit configurations in Field Programmable
Gate Arrays or
other semiconductor devices, or other non-transitory, tangible computer-
readable storage
media) encoded with one or more programs that, when executed on one or more
computers or
other processors, perform methods that implement the various embodiments of
the invention
discussed above. The computer readable medium or media can be transportable,
such that the
program or programs stored thereon can be loaded onto one or more different
computers or
other processors to implement various aspects of the present invention as
discussed above. As
used herein, the term "non-transitory computer-readable storage medium"
encompasses only a
computer-readable medium that can be considered to be a manufacture (i.e.,
article of
manufacture) or a machine and excludes transitory signals.
[00107] The terms "program" or "software" are used herein in a generic sense
to refer to any
type of computer code or set of computer-executable instructions that can be
employed to
program a computer or other processor to implement various aspects of the
present invention
as discussed above. Additionally, it should be appreciated that according to
one aspect of this
embodiment, one or more computer programs that when executed perform methods
of the
present invention need not reside on a single computer or processor, but may
be distributed in a
modular fashion amongst a number of different computers or processors to
implement various
aspects of the present invention.
[00108] Computer-executable instructions may be in many forms, such as program
modules,
executed by one or more computers or other devices. Generally, program modules
include
routines, programs, objects, components, data structures, etc. that perform
particular tasks or
implement particular abstract data types. Typically the functionality of the
program modules may
be combined or distributed as desired in various embodiments.
[00109] Also, data structures may be stored in computer-readable media in any
suitable form.
For simplicity of illustration, data structures may be shown to have fields
that are related through
location in the data structure. Such relationships may likewise be achieved by
assigning storage
for the fields with locations in a computer-readable medium that conveys
relationship between
DOCSTOR: 270846511 22
CA 02816469 2016-01-08
the fields. However, any suitable mechanism may be used to establish a
relationship between
information in fields of a data structure, including through the use of
pointers, tags, addresses or
other mechanisms that establish relationship between data elements.
[00110] Various aspects of the present invention may be used alone, in
combination, or in a
variety of arrangements not specifically discussed in the embodiments
described in the
foregoing and the concepts described herein are therefore not limited in their
application to the
details and arrangement of components set forth in the foregoing description
or illustrated in the
drawings. For example, aspects described in one embodiment may be combined in
any manner
with aspects described in other embodiments.
[00111] Also, the invention may be embodied as a method, of which several
examples haves
been provided. The acts performed as part of the method may be ordered in any
suitable way.
Accordingly, embodiments may be constructed in which acts are performed in an
order different
than illustrated, which may include performing some acts simultaneously, even
though shown
as sequential acts in illustrative embodiments.
[00112] While the invention has been described with reference to certain
exemplary
embodiments thereof, those skilled in the art may make various modifications
to the described
embodiments of the invention without departing from the true spirit and scope
of the invention.
The terms and descriptions used herein are set forth by way of illustration
only and not meant
as limitations. In particular, the present invention has been described by way
of examples of a
variety of devices and there may be other modifications or embodiments.
23
