科技產品生命週期之預測模型比較

Abstract

成長曲線模型常常被使用來預測科技產品之走向與趨勢，本研究利用22組科技產品資料來比較The simple logistic model, the Gompertz model, and the time-varying extended logistic model，此三種科技預測模型之預測準確度，再歸納出此三種模型之優缺點及建議使用時機。結果發現，The time-varying extended logistic model對在70%的科技產品上，都有比Simple logistic model 與Gompertz model兩個模型較好的預測準確度；但由於The time-varying extended logistic model在模型設定時需要有較多的參數來估計成長上限，在資料點太少的情況下，有約20%的機率無法得到收斂的結果，因此本研究建議若欲使用Extended logistic model，最好有15點以上之連續資料，且產品成長曲線有S曲線的軌跡，將會有較準確的預測結果。本研究亦提出一個選擇預測模型的決策流程，建議若在Extended logistic model無法收斂的情況下，但該產品成長曲線之反曲點已出現，則Simple logistic model 與Gompertz model可被使用來預測產品未來的發展空間。本研究亦利用大陸RFID專利申請案數量為一應用該決策流程之個案，並進一步預測未來RFID產業的發展趨勢。最後，本研究也提出對產品生命週期各階段之策略建議。

Many successful technology forecasting models have been developed but few researchers have explored a model that can best predict short product lifecycles. This research studies the forecast accuracy of long and short product lifecycle datasets using simple logistic, Gompertz, and the extended logistic models. Time series datasets for 22 electronic products were used to evaluate and compare the performance of the three models. The findings show that the time-varying extended logistic model fits short product lifecycle datasets 70% better than the simple logistic and the Gompertz models. A decision diagram is proposed to select a suitable forecasting model among the three models. The results suggest that there should be at less fifteen data points for the extended logistic model to reach better predictions. However, if the extended logistic model cannot be applied and the inflection point of the growth curve is revealed, the simple logistic and the Gompertz models can be the alternatives for forecasting the future trend of the product. A case study of China RFID patent forecast is also presented to demonstrate the selection procedure proposed in this research. Finally, the suggestions for product lifecycle management strategies in different lifecycle stages are also discussed.