TECHNOLOGY AREA(S): Materials/Processes

ACQUISITION PROGRAM: Strategic Systems Programs, ACAT I

OBJECTIVE: Develop innovative, predictive condition-based maintenance software to determine degradation and forecast production and refurbishment of hardware to reduce maintenance costs and increase operational availability.

DESCRIPTION: The acquisition program has an ongoing need to reduce total ownership costs and extend the life-cycle of components and systems to improve the reliability and overall operational readiness of the fleet. A cost effective method for ensuring component reliability is to augment the fixed schedule maintenance approach with deterministic component health and usage data to inform selective and targeted maintenance activities. The acquisition program seeks an innovative condition-based maintenance technology (i.e., a maintenance system that forecasts the health of the hardware) that can use adaptive learning techniques to “understand” component interdependencies and can accurately predict component failure of the system based on all available parametric data.

Innovative predictive software for forecasting the performance and maintenance of the hardware is required to address issues with the present method of maintenance. Currently, there is limited preventative maintenance that occurs on the hardware and is usually time-based and dependent upon human monitoring of systems. Hardware for this effort includes a two speed electric winch, wire rope, motor, brake, gearbox, and large metal structures. Computer-controlled test and monitor systems provide system status and allow for monitoring of key sub-system parameters such as fatigue, degradation, stress etc., but this data is not captured and thus not analyzed over time. Currently, preventative maintenance is not driven by automated system status or performance indicators and trends. Thus, maintenance is performed inefficiently and often fails to predict or prevent component and system failures.

Additionally, corrective hardware maintenance usually occurs after a component or system fails, or if component degradation is observed during routine maintenance. Failure to anticipate corrective maintenance requirements increases mean time to repair (MTTR), and decreases operational availability (Ao). Unanticipated corrective maintenance actions also drive up costs due to increased labor costs and expedited shipping costs when parts have to be obtained quickly.

Current software is not capable of making decisions but can be trained to improve its performance by factoring both technical decisions and programmatic decisions. An expert system that can use readily available, but not currently recorded, performance parameters to predict and thus preempt component and system failures is sought to improve overall system Ao, reduce MTTR, and reduce system maintenance and repair costs.
• Software should predict a failure, the inability or at least serious degradation of the hardware to perform its intended functions
o Software should determine the current degree of fault as quantified by Figure of Merit
o Prediction about the progression of the fault, in order to postulate the equipment’s degree of fault as a particular point in time in the future
o Determine the level of the fault, as quantified by the FOM that will produce a failure of the platform.
• Software will be able to use inputs from historical and manufacturing data along with data from current sensors on the equipment.
• Software should be able to predict the life of the equipment
• Software should provide the dynamic variation/ uncertainty boundaries on the prediction
• Software should be able to use both supervised and unsupervised learning
• Software can include but not limited to linear regression, linear multiple regression, time series analysis, Bayesian dynamic linear models and non-linear regress and multiple regress
A desired, innovative solution is needed to expertly and continuously monitor the component parametric data streams and conduct trend analysis. The expert system would combine the trend analysis data with component degradation and failure data reports to improve its prediction algorithms. The desired result is a system that is capable of providing a report such as, “hardware A” has a 90% probability of failure within the next 72-96 operating hours” or “the output of component “B” decreased by 10% in the last 7 days with the rate of output decrease accelerating significantly in the last 24 operating hours, indicating there is an 89% probability of component failure in the next 96 operating hours.”

PHASE I: Define and develop a predictive condition-based maintenance forecaster that meets the requirements described above and demonstrate the feasibility of the concept against hardware. Perform analysis, modeling and simulation, or laboratory investigations/demonstrations to provide initial assessment of approach feasibility.

PHASE II: Develop a prototype based on Phase I for evaluation. Validation of the software should include apparent, internal and external validation. Internal validation should include calibration with the data used to construct the predictive software, assessment of discrimination with the data and use of bootstrap to generate bias-corrected estimates of calibration and discrimination.

PHASE III DUAL USE APPLICATIONS: Perform assessments on the hardware using data collected from in-situ sensors, hardware manufacturers and historical data in order to provide a long range maintenance plan. Software predictions will be compared to actual degradation and life of the equipment. Extend the use of this predictive condition-based maintenance forecaster to additional hardware components through future required development. Private Sector Commercial Potential: A predictive maintenance forecaster would improve the operational reliability of all hardware and improve their availability. Commercial hardware manufacturers would be able to incorporate the technology into their sustained maintenance planning. This is an innovative capability that can be used in any industry that needs to increase operational availability (Ao) and mean time to repair (MTTR).

REFERENCES:

• Peng, Ying, Dong, Ming, and Zuo Jian M. Current Status of Machine Prognostics in Condition-based Maintenance. The International Journal of Advanced Manufacturing Technology, Volume 50, Issue 1, pages 297-313. 06 January 2010.
• Sun, Jianzhong, Zuo, Hongfu, Wang, Wenbin, and Pecht, Michael G. Application of a Stat Space Modeling Technique to System Prognostics based on a Health Index for Condition-Based Maintenance. Mechanical Systems and Signal Processing. Volume 28, pages 585-596. 29 November 2010.
• Voisin, A., Levrat, E., Cocheteux, P., Lung, B. Generic Prognosis Model For Proactive Mainteannce Decision Support: Application to Pre-Industrial E-Maintenance Test Bed. Journal of Intelligent Manufacturing. Volume 21, Issue 2, page 177-193. April 2010

KEYWORDS: Condition-based, maintenance, software, predictive, sustainment, sensors

** TOPIC AUTHOR (TPOC) **

DoD Notice:   Between April 22, 2016 and May 22, 2016 you may talk directly with the Topic Authors (TPOC) to ask technical questions about the topics. Their contact information is listed above. For reasons of competitive fairness, direct communication between proposers and topic authors is not allowed starting May 23, 2016 , when DoD begins accepting proposals for this solicitation.

However, proposers may still submit written questions about solicitation topics through the DoD's SBIR/STTR Interactive Topic Information System (SITIS), in which the questioner and respondent remain anonymous and all questions and answers are posted electronically for general viewing until the solicitation closes. All proposers are advised to monitor SITIS (16.2 Q&A) during the solicitation period for questions and answers, and other significant information, relevant to the SBIR 16.2 topic under which they are proposing.

If you have general questions about DoD SBIR program, please contact the DoD SBIR Help Desk at 800-348-0787 or sbirhelp@bytecubed.com