The concept of efficiency has always been and will continue to be important for competitive business environments where limited resources exist. Owing to the growing complexity of organizations and economy in general, this trend is expected to continue to remain a high priority for organizations. Continuous performance evaluations that utilize both qualitative and quantitative information play a significant role in sustaining efficient and effective business processes. Therefore, the literature offers a wide range of performance evaluation methodologies to assess the operational efficiency in various industries. Majority of these models, however, focus solely on quantitative criteria, avoiding the interrelations and dependencies between qualitative and quantitative measurements. Furthermore, these methodologies tend to utilize discrete and contemporary information eliminating historical performance data. With these motivations, this paper proposes an integrated approach combining fuzzy decision-making trial and evaluation laboratory (DEMATEL), analytic network process (ANP), and artificial neural network (ANN) methodologies for performance evaluation. In the proposed model, DEMATEL and ANP methodologies are utilized in a group decision-making concept to obtain priorities of the evaluation criteria. Following this, an ANN model is designed and trained with historical performance data collected from the organization and the results of the fuzzy DEMATEL-ANP model. The outcomes include the relational data among the criteria and alternatives used in the model in addition to their relative rankings. A food industry case study is presented to demonstrate the steps of the proposed model.

中文翻译：

---

一种用于绩效评估的具有偏好启发的智能多属性群体决策方法

对于资源有限的竞争性商业环境，效率的概念一直并将继续很重要。由于组织和经济总体上日益复杂，预计这一趋势将继续成为组织的重中之重。利用定性和定量信息的持续绩效评估在维持高效和有效的业务流程方面发挥着重要作用。因此，文献提供了广泛的绩效评估方法来评估各个行业的运营效率。然而，这些模型中的大多数只关注定量标准，避免了定性和定量测量之间的相互关系和依赖性。此外，这些方法倾向于利用离散的和当代的信息来消除历史性能数据。基于这些动机，本文提出了一种结合模糊决策试验和评估实验室 (DEMATEL)、分析网络过程 (ANP) 和人工神经网络 (ANN) 方法进行绩效评估的综合方法。在所提出的模型中，DEMATEL 和 ANP 方法被用于群体决策概念中，以获得评估标准的优先级。在此之后，使用从组织收集的历史绩效数据和模糊 DEMATEL-ANP 模型的结果设计和训练 ANN 模型。结果包括模型中使用的标准和替代方案之间的关系数据以及它们的相对排名。