Maintenance optimization for asset performance management

• US 10,466,689 B2
• Filed: 04/03/2017
• Issued: 11/05/2019
• Est. Priority Date: 09/22/2015
• Status: Active Grant

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Adendorff, Michael, Tiozzo, Fabio
• Primary Examiner(s): Cook, Brian S
• Application Number:
  US15/477,158
• Publication Number:
  US 20170205818A1
• Time in Patent Office: 946 Days
• Field of Search: 703 7
• CPC Class Codes: G05B 23/0248 Causal models, e.g. fault tree; digraphs; qualitative physics
• CPC Class Codes: G05B 23/0278 Qualitative, e.g. if-then rules; Fuzzy logic; Lookup tables; Symptomatic search; FMEA
• CPC Class Codes: G05B 23/0281 Quantitative, e.g. mathematical distance; Clustering; Neural networks; Statistical analysis
• CPC Class Codes: G05B 23/0283 Predictive maintenance, e.g. involving the monitoring of a system and, based on the monitoring results, taking decisions on the maintenance schedule of the monitored system; Estimating remaining useful life [RUL] preventive maintenance, i.e. planning maintenance according to the available resources without monitoring the system G06Q10/06
• CPC Class Codes: G06F 11/0793 Remedial or corrective actions recovery from an exception in an instruction pipeline G06F9/3861; by retry G06F11/1402; for recovering from a failure of a protocol instance or entity H04L69/40
• CPC Class Codes: G06N 7/01 Probabilistic graphical models, e.g. probabilistic networks
• CPC Class Codes: G06Q 10/0635 Risk analysis of enterprise or organisation activities
• CPC Class Codes: G06Q 10/20 Administration of product repair or maintenance
• CPC Class Codes: G07C 5/006 Indicating maintenance
• CPC Class Codes: Y02P 90/80 Management or planning

Abstract

A computer implemented method comprising receiving one or more predictive maintenance models each defining a time-based probability of failure for one or more components, receiving current performance data for the components, defining a failure function for each component from a predictive maintenance model for the component and the current performance data for the component, the failure function defining the probability of failure of the component in each of a set of time periods, defining a value loss function for each component from the failure function for the component and a time-based component cost, the value loss function defining the expected value loss due to a planned replacement of the component in a given time period before the component fails or reaches its scheduled end-of-life, receiving data defining one or more factors that have an impact on the cost of a maintenance option.

Claims

1. A computer-implemented method comprising:

a processor capable of performing a method comprising;

receiving one or more predictive maintenance models each defining a time-based probability of failure for one or more components,

receiving current performance data for the components,

defining a failure function for each component from a predictive maintenance model for the component and the current performance data for the component, the failure function defining the probability of failure of the component in each of a set of time periods,

defining a value loss function for each component from the failure function for the component and a time-based component cost, the value loss function defining the expected value loss due to a planned replacement of the component in a given time period before the component fails or reaches its scheduled end-of-life,

receiving data defining one or more factors that have an impact on the cost of a maintenance option,

defining one or more time buckets, wherein the one or more time buckets define a time period in which maintenance takes place for the one or more components, and wherein maintenance plans for the one or more components are optimized in response to the maintenance for two or more components being planned in a same time bucket associated with the one or more time buckets,

for each of the one or more time buckets, determining a machine downtime due to an unexpected failure of the one or more components, and determining an expected impact on production and on maintenance resources due to the machine downtime,

creating an optimization model for all components from the component failure functions, the value loss functions and the data defining one or more factors that have an impact on the cost of a maintenance option, the solutions to the optimization model representing the per time period effect of one or more maintenance options.

2. The computer-implemented method according to claim 1, further comprising:

receiving data defining a maximum maintenance capacity per time period for all components and when creating the optimization model for all components to define maintenance options that do not exceed the maximum maintenance capacity per time period for all components.

3. The computer-implemented method according to claim 1, further comprising:

receiving an input defining a set of desired optimization outcomes, each defined optimization outcome defining one or more factors to be optimized and wherein creating the optimization model for all components defines maintenance options that each provide optimization of each factor defined in a desired optimization outcome.

4. The computer-implemented method according to claim 1, further comprising:

determining a preferred maintenance option with respect to a desired optimization outcome and output an illustration of the preferred maintenance option as a matrix of components against time periods plus the impact of the failure function for each component as a component cost.

5. The computer-implemented method according to claim 4, further comprising:

outputting a selection of maintenance options and receive an input defining a preferred maintenance option.