EP1069487A1
Expired / Lapsed
EP
⏱️ 0.0 years remaining
Automated diagnosis of printer systems using bayesian networks
Assignee: HP Inc
•
Filing: 2000-07-03
•
Granted: —
•
Expires: 2020-07-03
Abstract
Abstract An automated troubleshooter uses Bayesian networks to troubleshoot a system (209,210). Knowledge acquisition is performed in preparation to troubleshoot the system (209,210). An issue to troubleshoot is identified. Causes of the issue are identified. Subcauses of the causes are identified. Troubleshooting steps are identified. Troubleshooting steps are matched to causes and subcauses. Probabilities for the causes and the subcauses identified are estimated. Probabilities for actions and questions set are estimated. Costs for actions and questions are estimated.
Inventors
- 👤 Claus Skaanning
- 👤 Finn V. Jensen
- 👤 Uffe Kjaerulff
- 👤 Paul A. Pelletier
- 👤 Lasse Rostrup Jensen
- 👤 Marilyn A. Parker
- 👤 Janice L. Bogorad
Bibliographic Details
Application Number
—
Jurisdiction
EP
Data Source
google_patents
Last Synchronized
2026-10-24 15:42
CPC Classifications
G; G05; G05B; G05B23/00; G05B23/02; G05B23/0205; G05B23/0259; G05B23/0275; G05B23/0281
Claims Analysis
Total Claims: 10
Claim 1
Independent Claim
127 words
A Bayesian network (1-11,500-531,601-608,610-629) that models a
system component (209,210) causing failure of a system, the Bayesian
network (1-11,500-531,601-608,610-629) comprising: an indicator node (502-512) that has a state that indicates whether the
system component (209,210) is causing a failure; a plurality of cause nodes (602-605,611-620), coupled to the indicator
node (502-512), each cause node in the plurality of cause nodes (602-605,611-620)
representing a cause of the system component (209,210) producing a
failure; and, a first plurality of troubleshooting nodes (606-608,621-627), each
troubleshooting node representing a troubleshooting step, each
troubleshooting node being coupled to at least one cause node from the
plurality of cause nodes (602-605,611-620), each troubleshooting step
suggesting an action to remedy causes represented by any cause nodes (602-605,611-620)
to which the troubleshooting node is coupled.
Claim 2
Dependent Claim
64 words
A Bayesian network (1-11,500-531,601-608,610-629) as set out in
claim 1, additionally comprising: a question node (628,629), the question node (628,629) being coupled to
at least one cause node from the plurality of cause nodes (602-605,611-620),
the question node (628,629) representing a question, which when answered,
provides potential information about causes represented by any cause nodes
(602-605,611-620) to which the question node (628,629) is coupled.
Claim 3
Dependent Claim
42 words
A Bayesian network (1-11,500-531,601-608,610-629) as set out in
claim 1, wherein the plurality of cause nodes (602-605,611-620) are coupled to
the indicator node (502-512) through a causes node which represents a
probability distribution over causes for failure of the system component
(209,210).
Claim 4
Dependent Claim
56 words
A Bayesian network (1-11,500-531,601-608,610-629) as set out in
claim 3, additionally comprising: a question node (628), the question node (628) being coupled to the
causes node (610), the question node (628) representing a question of a
general type which is not necessarily related to a symptom or a cause of
failure of the system component (209,210).
Claim 5
Dependent Claim
50 words
A Bayesian network (1-11,500-531,601-608,610-629) as set out in
claim 1 wherein for a first troubleshooting step suggesting a first action, when
calculating whether the first action will solve a first cause, utilizing an
inaccuracy factor, the inaccuracy factor representing a probability that a user
will incorrectly perform the first action.
Claim 6
Independent Claim
92 words
A method for performing knowledge acquisition used to troubleshoot
a system (209,210), the method comprising the following steps: (a) identifying an issue to troubleshoot; (b) identifying causes of the issue; (c) identifying subcauses of the causes; (d) identifying troubleshooting steps; (e) matching troubleshooting steps to causes and subcauses; (f) estimating probabilities for the causes identified in step (b) and the
subcauses identified in step (c); (g) estimating probabilities for actions and questions set out in the
troubleshooting steps; and, (h) estimating costs for actions and questions set out in the
troubleshooting steps.
Claim 7
Dependent Claim
41 words
A method as in claim 6 wherein in step (d) each troubleshooting step
involves an action in one of the following categories: (a) a solution action that provides an action that may solve the issue;
and, (b) an information gathering action.
Claim 8
Dependent Claim
49 words
A method as in claim 6 wherein in step (e) each troubleshooting step
that includes an action is matched to any cause or subcause the action can
solve and each troubleshooting step that includes a question is matched to
any cause or subcause to which the question is related.
Claim 9
Dependent Claim
82 words
A method as in claim 6 wherein step (g) comprises the following
substeps: (g.1) determining for each related cause or related subcause a first
probability that indicates whether performing a first action correctly will
solve the issue; (g.2) determining a second probability that indicates a likelihood a
customer will perform the first action correctly; and (g.3) combining the first probability and the second probability to
determine whether instructing the customer to perform the first action will
result in the issue being solved.
Claim 10
Dependent Claim
66 words
A method as in claim 6 wherein in step (h), costs for a first action
include factors that take into account the following: time to perform the first action; risk of breaking something when performing the first action; amount of money required to purchase any parts necessary to perform
the first action; and, degree of insult a user may experience when the first action is
suggested.
References Cited (Backward Citations)
No citation records indexed.
USPTO Annuity & Maintenance Windows
Maintenance schedule is calculated for granted US utility patents.