考慮使用正副廠零件與季節性因素下機車備用零件需求預測之研究

A study on Demand Forecasting of Motorcycle Spare Parts: Considering the Usage Rate of Genuine Parts and Seasonal Factors

賴以修(Yi-Siou Lai)

指導教授 ： 張淳智

國立臺中科技大學/管理學群/流通管理系碩士班/碩士(2011年)

摘要

機車產業在目前經營環境不斷改變下，若備用零件之庫存能良好的管控，將可使機車維修服務的品質得以維持，其中備用零件需求預測的精確度的好壞便顯得格外重要，然而，備用零件之需求量為隨機且不規則發生的需求型態，此讓事前作需求預測較為困難，本研究針對機車產業備用零件需求預測議題加以探討，盼能找出最適宜之備用零件需求預測方法。
本研究模式建構上考慮到使用正副廠零件與季節性因素，並以貝氏統計配合馬可夫鏈蒙地卡羅法對模式參數進行校估。實證研究主要挑選出40種備用零件，並比較本研究方法實證結果與時間序列預測方法(移動平均法、指數平滑法、CR法與各CR法修正法)以及常態壽齡模式的需求預測能力。
結果發現，本研究模式之預測效果最佳，代表本研究模式在預測間歇型需求資料上是相當適切的。此外，值得一提的是本研究實證結果中使用正廠零件比率之資訊，可供機車產業管理者作為訂定經營策略之用。

關鍵字

需求預測 ； 備用零件 ； 馬可夫鏈蒙地卡羅法 ； 正廠零件

並列摘要

The environment of motorcycle industry is changing currently. If the spare parts in inventory can be control well, it will make the motorcycle maintain service to keep the quality. Among them, the spare parts demand forecasting accuracy will be particularly important. Therefore, the random demand and uncertainty for spare parts make the demand forecasting more difficult. This study discuss this problem and hope to find the most appropriate method to forecast demand for spare parts.
This research considers to introduce the usage rate of genuine spare parts and seasonal factors to construct the demand models, and we through Bayesian statistics with Markov Chain Monte Carlo method (MCMC) to estimate model parameters. This research choose 40 spare parts, and Comparing the forecasting abilities of our models and time series forecasting methods(moving average method, exponential smoothing method, CR method, and the modification CR method), and the normal life time model.
As a result, our models have the best performance for demand forecasting, and our models are quite appropriate for forecasting intermittent demand. In addition, using the information of genuine spare parts ratio in this research, can help the motorcycle industry managers to adjust their market strategy.

並列關鍵字

Demand Forecasting ； Spare Parts ； Markov Chain Monte Carlo ； Genuine Parts

參考文獻

1.吳沛軒 (民97)。考量零件生命週期下之汽車售後零組件需求預測與備貨模式。國立台灣大學商學研究所碩士論文，未出版，臺北市。

5.張瀚陽 (民98)。馬可夫鏈蒙地卡羅法於機車零備件需求預測之研究。國立台中技術學院流通管理系碩士論文，未出版，臺中市。

1.Aronis, K. P., Magou, I., Dekker, R., & Tagaras, G. (2004). Inventory Control of Spare Parts Using A Bayesian Approach: A Case Study. International Journal of Forecasting 20, 375-387.

3.Bhat, C. R. (2001). Qusai-random Maximum Simulated Likelihood Estimation of the Mixed Multinomial Logit Model. Transportation Research B 35, 677-693.

4.Bhat, C. R. (2003). Simulation Estimation of Mixed Discrete Choice Models Using Randomized and Scrambled Halton Sequences. Transportation Research B 37, 837-855.