Grading of Raw Jute Fibres Using Criteria Importance through Intercriteria Correlation (CRITIC) and Range of Value (ROV) Approach of Multi-criteria Decision Making

ABSTRACT

In modern BIS system, a scientific approach is adopted for grading the raw jute fibers based on six apposite physical properties, namely fiber strength, defect, root content, fineness, color, and bulk density, of which the latter two parameters are still evaluated subjectively. Several researchers have attempted, over the years, many exponents of multi-criteria decision making (MCDM) approaches and got encouraging results. However, in all these approaches, subjective weighting methods have been used in the form of analytic hierarchy process (AHP). In this paper, a maiden integrated approach using criteria importance through intercriteria correlation (CRITIC) and range of value (ROV) is presented for grading jute fibers on the basis of abovementioned six attributes. Ranking performance of the present approach closely resembles with other approaches hitherto attempted by several researchers. The efficacy and robustness of the new approach are validated using sensitivity analysis through altering the weightage levels of the criteria. The rank correlation coefficients between the ranking patterns obtained in each of the weightage levels are, in most of the cases, 1 or very close to 1, which signifies no occurrence of rank reversal or change in ranking. During sensitivity analysis in dynamic decision conditions, the approach is stable up to certain level and then rank reversal takes place for some of the alternatives including the top ranked one. The uniqueness of the present study lies in the fact that here criteria weights are determined objectively using CRITIC and hence more reliable and devoid of influence/preferences of decision-makers (DMs). Moreover, the new approach is very simple with a few simple mathematical equations, albeit a potent tool of decision making with strong background logic.

摘要

现代BIS系统采用科学的方法，根据黄麻纤维强度、疵点、根含量、细度、色泽、容重等6个相对应的物理性质对黄麻原纤维进行分级，并对后两个参数进行主观评价. 多年来，许多研究者尝试了多准则决策方法，并取得了令人鼓舞的成果. 然而，在所有这些方法中，主观加权方法都是以层次分析法（AHP）的形式使用的. 本文提出了一种基于CRITIC和ROV的黄麻纤维分级方法。该方法的排序性能与一些研究人员迄今为止尝试的其他方法非常相似. 通过改变标准的权重水平，通过灵敏度分析验证了新方法的有效性和鲁棒性. 在大多数情况下，在每个权重级别中获得的排名模式之间的排名相关系数是1或非常接近1，这表示排名中没有出现排名反转或变化. 在动态决策条件下进行灵敏度分析时，该方法稳定到一定水平，然后对一些方案（包括排名靠前的方案）进行秩反转. 本研究的独特之处在于，这里的标准权重是使用批评家客观地确定的，因此更可靠，没有决策者的影响/偏好. 此外，新方法非常简单，只需几个简单的数学方程，尽管它是一个强大的决策工具，具有很强的背景逻辑.

关键词:

• 黄麻
• 分级

KEYWORDS:

• CRITIC
• jute
• grading
• MCDM
• ROV