| 摘要: | 隨著世界各國武器科技的進步及作戰型態的改變，從現今作戰場景與戰術思維的觀點來看，没有空防即没有所謂的國防，防空系統已成為主導戰場優勢的關鍵性武器之一，然而防空戰備的能力得以有效的發揮，則端賴平時裝備妥善的維持與良好後勤維保的運作，方能捍衛領空，殲敵於千里之外；惟現行國軍針對國防裝備所制定的維保模式，雖沿習美軍後維制度採行O(單位段)、I(野戰段)、D(基地段)三段五級的維修階層策略，依照過去維修經驗常數，制定各項裝備定更、定檢件的更換頻次與時程，並以系統使用時數為基準的預防保養制度（Time Based Preventive Maintenance, TBPM），然而對於架構設計複雜且精密的防空系統而言，若無法有效預測各系統元件失效的資訊，進行預防性修護的作業，恐難以維持系統妥善，進而影響部隊戰力及作戰任務的遂行。本研究以國軍現役短程防空系統偵測模組總成歷史維修及耗用料紀錄為基礎，透過資料科學分析方法中，描述性統計（Descriptive Statistics）、單因子變異數分析（One-way ANOVA）、皮爾森積差相關分析（Pearson Product-Moment Correlation）及時間序列分析（Time Series Analysis）等技術，分析各元件間失效頻次、關聯性及優序性之成因，再藉由美軍武器系統可靠度分析理論，擷取總成項下部分元件之串並聯網絡進行可靠度數據分析，並探討其後勤效益性與最佳優化模式，據以對國軍各層級維修廠庫及維保人員，在修護作業實務上，提出有效的預防維修模式，以期降低系統重複拆解的次數及工時，並且提昇整體系統可靠度。;Regarding today`s battle situation and tactical concept, the development of weapon technologies and changes of operational patterns in the world have resulted in a conclusion of “No Air Defense, No National Defense”. Air Defense System has become one of the critically advantageous weapon systems to lead the battlefield. In fact, air defense capabilities rely on the availability rate maintaining and logistics managing, for the purpose of guarding the air space and destroying the enemy thousands of miles away. The current maintenance module for national defense equipment of R.O.C. Armed Forces was established in accordance with logistical maintenance mechanism of U.S. Army. That provides for three levels - Organizational, Intermediate and Depot - and five echelons of maintenance categories, developing the replacing frequency and timeline for periodical examining items of materiel and following the utility-hour-based Time Based Preventive Maintenance (TPBM). However, for the complicatedly and precisely structural designed Air Defense System, it is hard to maintain the availability rate if failing to forecast information on malfunctioned items of systems, and to execute preventive maintaining operations. It will also significantly affect on the war fighting functions and on combat tasks execution. This research is based upon maintenance and repairing history and material consumption record of active Short Range Air Defense Systems detection module assembly, and using data science analysis, for instances Description Statistics, One-way ANOVA, Pearson Product-Moment Correlation, and Time Series Analysis, to analyze malfunction frequencies among items, their connections and priorities of causes. After that, according to the reliability analysis theory of the U.S. military weapon system, the research capture part string of assembly items and link the internet conducting reliability data analysis. Then discuss logistical efficiency and optimization model. Finally, the expectation of this research is to provide preventive maintenance standards on practices for maintainers in levels and echelons of repairing and maintenance shops for escalating system reliability through lowering repeatedly dissembling frequency and its working hours. |
