本文嘗試採用資料包絡分析法(DEA)來評估基隆港三個貨櫃基地歷年之生產效率。 藉由效率分析、差額分析及敏感度分析，說明投入要素與產出組合未充分運用之程度，以及 造成生產無效率的原因。分析結果發現，基隆港第三貨櫃基地平均生產效率最高 (平均效率 值為 0.9529)，第一貨櫃基地次之 (0.9243)， 第二貨櫃基地因後線作業量較少，效率最低 (0.8171)。能源、橋式機、後線機具等資源未能充分運用，係造成各貨櫃基地生產無效率的 主要原因。本文建議，短期內基隆港各貨櫃基地應加強機具之維修保養；第一貨櫃基地應加 強橋式機的維修汰換，惟其總數可維持不變；第二貨櫃基地之後線機具可考慮出租或售予航 商；第一貨櫃基地與第三貨櫃基地間之船舶修造廠應予以遷移，將兩貨櫃基地合併，俾利後 線場地、機具互相支援，以及碼頭工人靈活調用。長期而言，未來貨櫃基地之棧埠作業宜朝 民營化方向的努力，可考慮由碼頭工人自組裝公司進行裝卸作業，以提昇生產效率。

　　　　　This paper attempts to evaluate the productive efficiency of Keelung Port container terminals by using Data Envelopment Analysis (DEA) technique. The causes and extent of inefficiency for input factors and output combinations are explained via efficiency analysis, slack analysis, and sensitivity analysis. The results indicate that terminal No.3 has the highest average efficiency (0.9529), followed by terminal No. 1 (0.9243). Terminal No. 2 has the lowest average efficiency (0.8171) due mainly to insufficient yard handling activities. The results also show that energy consumption, quay cranes, and yard machines are the major factors responsible for productive inefficiency. To improve the container handling efficiency, we suggest that Keelung Port Authority enhance the maintenance of container equipment, sell or rent yard equipment to the yard leaseholder for terminal No. 2, relocate the Ship & Machinery Works between terminals No. 1 & No. 3 so that both yards can share the equipment and stevedores. Finally, privatization of terminal operations by encouraging stevedores to form a container handling company seems a promising direction in enhancing the productive efficiency.