灰色理論預測台灣辦公大樓租戶和公共區域修繕時間

Utilizing Grey Theory for Predicting Maintenance Hours for Office Building Tenants and Public Areas in Taiwan

本研究以高層辦公大樓為例，利用灰色理論研究辦公室和公共區域設施設備的回應性維修服務。目的是為提供辦公大樓業主、執行經理、物業管理公司、機電公司的決策和評估預測模型參考。
本研究的主要發現總結如下：
1.主要探討電力系統、建築系統、供水系統、空調系統、消防系統和人力調度六類設施維修的重要性。設施維護的重要性順序如下：電力系統、建築系統、供水系統、空調系統、人力調度和消防系統。
2.本研究採用GM(1,N)及迴歸方法，以維修時間為依變數，以維修次數、租賃面積、租戶數為自變數。根據單年1月至12月的12個數據點進行了每月預測。驗證結果表明，GM(1,N)方法的平均絕對誤差為6.41%，平均準確度為93.59%，而迴歸模型的平均絕對誤差為4.66%，平均準確度為95.34%。這些發現表明這兩種方法都具有很高的預測準確性。

This study focused on case office buildings and investigated the responsive repair requests for facilities and equipment in offices and public areas using gray theory. The purpose was to provide valuable insights for future office building owners, executive managers, property management companies, and mechanical and electrical companies in making decisions and assessing forecast models. The key findings of this study are summarized as follows:
1. Grey Relational Analysis examined the importance of facilities repair in six categories: power systems, building systems, water systems, air conditioning systems, fire systems, and manpower dispatch. The order of importance for facilities maintenance was identified as follows: power systems, building systems, water supply systems, air conditioning systems, manpower dispatch, and firefighting systems.
2. By employing the GM (1, N) and regression methods, this study utilized maintenance hours as the dependent variable and repair number, leased area, and number of tenants as independent variables. It conducted monthly forecasts based on 12 data points from January to December specific on year Four. The verification results demonstrated that the GM (1, N) method had a mean absolute error of 6.41% and an average accuracy of 93.59%, while the regression model had a mean absolute error of 4.66% and an average accuracy of 95.34%. These findings indicate that both methods possess a high level of accuracy in forecasting.