以基於DDANP-mV之品質機能展開法定義汽車零組件生產之精實策略

Abstract

汽車零組件產業為汽車工業發展之重要推手，近年來，由於需求波動劇烈，不易預測，而且多數訂單少量多樣，客戶且不斷嘗試壓低採購價格，對零組件廠商造成極大壓力，因此需要異於傳統的生產方式。為因應潮流，精實生產逐漸盛行，以期使用最低成本，達到最佳品質。雖然將精實生產方式導入零組件生產線，極為重要，但是相關研究甚少，引此本研究擬定義一基於多準則決策模式之分析架構，整合品質機能展開法，分析客戶需要，及其與零件規格之關聯性之後，以萃思法推衍適合之發明策略，改善產線。首先，本研究依據實際狀況，歸納客戶需要，並且將差速器之規格，導入品質屋。其次，本研究以決策實驗室法 (Decision Making Trial And Evaluation Laboratory，DEMATEL)推衍客戶需要間之影響關係，並以決策實驗室法為基礎之網路流程 (DEMATEL based Analytic Network Process，DANP)，計算各需要之權重，其後，以折衷排序法 (VlseKriterijuska Optimizacija I Komoromisno Resenje，VIKOR)，邀集專家，計算需要優先改善之規格參數。最後以萃思法，針對衝突嚴重之優先參數，訂定精實生產策略。本研究以台灣某上市汽車零組件公司之差速器為個案，邀集專家，針對客戶臨時變更訂單，要求大幅增加產量，並降低採購單價之特殊情境，以所發展之分析架構，訂定精實策略。本研究將精實生產原則應用於汽車差速器生產線，以達提高產能並降低生產成本且能維持生產良率，並達成客戶需求。以本決策架構選出，適合個案分析中差速器之法精實策略包括「全面預防維護」 (Total Preventive Maintenance， TPM)、「標準作業」(Standard Work)和面向製造和裝配的設計 (Design for Manufacturing and Assembly， DFMA)為最適合用於個案差速器之精實策略。研究結果顯示，混合多準則分析模型用於精實生產的可行性。完整驗證之架構，也可作為其他產品訂定精實生產策略之用。

The automotive component industry is an important driving force for the development of the automotive industry. In recent years, due to fluctuations in demand and small and diverse orders of most automotive components, accurate predictions of demands have been difficult. Customers are constantly trying to negotiate for lower purchase prices, which has caused great pressure on component manufacturers. Therefore, a different approach from the traditional ways of manufacturing automotive components is required. In response to the trend, lean production strategies have widely been adopted to achieve the best quality with the lowest cost. Although the introduction of lean production methods into the production lines of automotive components is extremely important, there are few related studies. Therefore, this study aims to define an analysis framework based on the quality function deployment (QFD) method with multi-criteria decision-making frameworks. Based on customer needs and the parameters of the automotive component, appropriate lean manufacturing strategies can be derived using the TRIZ method.First, this research summarizes the needs of customers based on an actual scenario, and it introduces the specifications of differential gear into the House of Quality (HOQ). Second, this research uses the Decision Making Trial and Evaluation Laboratory (DEMATEL) to derive the influence relationship between customer needs, and it uses the MATEL-based Analytic Network Process (DANP) as the basis to derive the weight versus each need. Then, the VlseKriterijuska Optimizacija I Komoromisno Resenje (VIKOR) method is introduced to derive the parameters that should be improved first. In the end, lean production strategies are formulated according to the priority parameters with serious conflicts using the TRIZ method.This research takes the differential gear of a listed automobile component company in Taiwan as the empirical case. Experts are invited to provide opinions regarding the special scenario of temporarily changing orders, greatly improved capacity, and lowering the unit cost of purchasing. The proposed analytic framework is used for decision-making. This research applies the principle of lean production to the production line of automobile differentials to increase production capacity and reduce production costs, maintain production yield, and meet customer needs. Based on the analytic results, differential gear total preventive maintenance (TPM) and standard work and design for manufacturing and assembly (DFMA) are the most suitable lean strategies for the empirical case. The research results demonstrate the feasibility of the proposed hybrid multi-criteria analysis model for lean production. The fully verified analytic framework can also be used for formulating lean production strategies of other components.