This study aims to cluster Turkish cities based on their local agricultural production and rank them in terms of performance by combining cluster analysis and multi-criteria decision-making (MCDM) methods. In this context, a three-phase methodology is developed. In the first phase, Ward's method is utilized to cluster cities according to agricultural production characteristics. In the second phase, the objective criteria weights are determined using the Criteria Importance Through Intercriteria Correlation technique (CRITIC). In the third phase, to rank the clusters in terms of performance, the Technique for Order Preference by Similarity to Ideal Solution (TOPSIS) method is applied. Due to the results, the 81 cities are divided into six clusters in terms of agricultural production features. The cluster with the highest performance is Cluster 6, in which Konya is alone. Cluster 4, which includes Antalya and Mersin, follows this cluster. Cluster 1 with 25 cities and Cluster 2 with 19 cities are the clusters with the lowest results. The results show that only a few cities such as Konya, Antalya, and Mersin are generating more than tens of them in combination. These findings reveal that local governments should reconsider their agricultural programs and develop new strategies under the direction of the central government.

中文翻译：

---

将聚类分析与 MCDM 方法相结合，用于评估当地农业生产

本研究旨在根据当地农业生产对土耳其城市进行聚类，并通过结合聚类分析和多标准决策 (MCDM) 方法对它们进行绩效排名。在此背景下，开发了一种三阶段方法。第一阶段，根据农业生产特点，利用沃德方法对城市进行聚类。在第二阶段，客观标准权重通过标准间相关性技术 (CRITIC) 确定标准重要性。在第三阶段，为了根据性能对集群进行排名，应用了与理想解相似度排序偏好技术 (TOPSIS) 方法。由于结果，81个城市在农业生产特征方面被划分为六个集群。性能最高的集群是集群 6，其中只有 Konya。包括安塔利亚和梅尔辛的集群 4 紧随该集群。包含 25 个城市的集群 1 和包含 19 个城市的集群 2 是结果最低的集群。结果表明，只有科尼亚、安塔利亚和梅尔辛等少数几个城市产生了几十个以上的组合。这些调查结果表明，地方政府应重新考虑其农业计划并在中央政府的指导下制定新的战略。