品質成本是指與品質相關活動的總成本，是衡量品質管理績效的一項重要指標，儘管有許多研究文獻強調了營建工程專案中品質成本的重要性，但國內卻鮮有文獻探討品質成本模型於營建工程中的相關應用。而品質成本之歷史資料多具備時間序列資料之特性，然而過往之研究受限於研究方法，僅能呈現已經發生之品質事實，無法詮釋具時序特性之成本資料背後所蘊含之意義。因此本研究旨在建立以時間序列分析技術輔助之品質成本估算方法，並透過蒐集國內外時間序列之文獻、選擇適當的時間序列模型等方法來建立品質成本推估模型，同時以先前研究所採用之柔性鋪面工程案例品質成本資料進行實際案例分析。研究結果顯示，本研究所建立之模型對該案例之預測效能尚可，但其結果不夠穩定，需要後續研究利用其餘時間序列品質成本資料進行驗證並提高模型預測準確度；此外於模型分析過程中輸出之趨勢與季節性之視覺化結果可以讓使用者清楚了解其品質成本走向的變化與各年的季節性變動，透過與先前研究之成果相比，證實了該視覺化結果的正確性，而且對整體走向的呈現與產出效率都優於先前研究之成果。;The cost of quality is defined as the total cost of quality-related activities and serves as an important indicator of quality management performance. Despite the prevalence of studies emphasizing the significance of the cost of quality in construction projects, there is a paucity of literature exploring the application of quality cost modeling in construction projects. Moreover, the historical data of quality costs are often characterized by time series data. However, previous studies are constrained by the research methodology, which is unable to present the underlying meanings behind the cost data with time series characteristics. Consequently, the objective of this study is to develop a quality cost estimation method based on time series analysis and to construct a quality cost estimation model by collating domestic and foreign time series literature, selecting appropriate time series models, and conducting a case study on the quality cost data of flexible paving projects utilised in the previous study. The findings of this study indicate that the model developed in this study remains effective in predicting the quality cost of this case. However, the results are not sufficiently stable, and further verification is required through the use of other time series quality cost data in subsequent studies to enhance the accuracy of the model′s prediction. Moreover, The visualization results of the trend and seasonality output in the process of the model analysis facilitate the users′ comprehension of the changes in the direction of the quality cost and the seasonal changes of the year. These results can be compared with those of the previous study to confirm the quality cost estimation model. A comparison of the results with those of the previous study demonstrates that the visualization results are accurate and that the overall trend and output efficiency are superior to those of the previous study.
