Predictive Analytics for Ship Scheduling
Abstract
Periodic maintenance and modernization of naval vessels are conducted in the shipyards. According to a recent article published by the U.S. Government Accountability Office (GAO), "The Navy s four shipyards completed 38 of 51 (75 percent) maintenance periods late" during fiscal years 2015 to 2019, causing late return of the vessels to the fleet. A naval vessel s late return to the fleet results in a decrease in operational readiness due to the reduced number of operational days available for these vessels and impacts on other vessels assigned to support the same or a complementary mission. There exists more than one interdependent factor that contributes to scheduling complexities. These include: inadequate planning for resources, quantity of overtime labor, direct yard costs, and quantity of work stoppages experienced that contribute to the tardiness in availability. Clearly, an effective yard scheduling process in NAVSEA requires careful consideration of multiple objectives and many metrics provided by different entities (32 SEA 21/CNRMC), and analysis of data that are voluminous, come at different velocity, and consist of a different variety or types. Data required for yard scheduling in NAVSEA comes from multiple sources and organizations, a few of them are listed below: Data sources: NMD, VSB, RMC/100-800, TAAS-INFO, eDFS, CASREPS, Spreadsheets, Shortage Reports, TSRA, Briefings, Port workload forecasts Data owners: NSWC-Corona, CNRMC 400/700, SURFMEPP, CNSP, CNSL, Amphib PAPM, FSC, etc. Moreover, the collected data is of various types, such as structured data, Time series, Text (eventually may involve images). Data may be sparse and contain data of different quality, and the data sources are sampled at different rates (daily, weekly, monthly, quarterly, or on an ad-hoc basis). Enabling predictive analytics for the yard scheduling problem must go through a full data science pipeline that contains extensive upstream tasks for data conditioning and assessment before being subjected to advanced analytics and modeling. The goal of this project is to enable upstream data science capabilities that aids downstream AI/ML or predictive analytics to quantitatively achieve CNO availability process milestones. Up until now, PASS research activities have focused on building predictive models of avails during execution with the goal of quantifying delay by enabling cost and schedule mitigation. The NJIT research team has successfully achieved the outlined goals. The next planned activities in PASS will focus on prediction in planning. Prediction in planning is about identifying the risk and quantifying the impact by enabling right size packaging for facilitating mitigation. The goal of these activities is to build models that predict how many days an avail needs in a yard
Document Details
- Document Type
- DoD Grant Award
- Publication Date
- Nov 08, 2024
- Source ID
- N000142412466
Entities
People
- Senjuti Basu Roy
Organizations
- New Jersey Institute of Technology
- Office of Naval Research
- United States Navy