TY - CONF

T1 - Prioritization of Alternatives with AHP PlusResponse Latency and Web Surveys

AU - Errore, Anna

AU - Lombardo, Alberto

AU - Barone, Stefano

PY - 2013

Y1 - 2013

N2 - Purpose: Overcome some drawbacks of the widely applied Analytical Hierarchy Process(AHP) related to the use of a rating scale for respondent’s judgments and propose a new methodto prioritize alternatives based on web surveys.Methodology/Approach: We use the response latency, i.e. the time taken to make a decision,to replace the rating scale in the pairwise comparisons which is a fundamental part of the AHPprocedure. A method combining the AHP with a mathematical model for the response latencyis proposed in the paper.Findings: Eliminating the rating scale and leaving only the “take the best” task alleviates therespondent and makes the survey much faster and intuitive. Moreover the results look morereliable in terms of consistency of judgments.Research Limitation/implication: An application to the development of a new service is usedto test the proposed method. Good results of the method are achieved also thanks to the use ofweb survey platforms, today easily accessible to both researchers and respondents; strengthsand weaknesses of web surveys are taken into consideration.Originality/Value of paper: The integration of the response latency in the traditional procedureof AHP gives two interesting contributions: on one hand it can be seen as a way to overcome,or at least alleviate, some limitations of the AHP; on the other hand it is a natural wayto involve the response latency, which is a measure quite well known in many scientific fields but seldom modeled and very rarely utilized in managerial decisions. Nowadays with the use ofweb surveys that measure can be much more easily manageable.

AB - Purpose: Overcome some drawbacks of the widely applied Analytical Hierarchy Process(AHP) related to the use of a rating scale for respondent’s judgments and propose a new methodto prioritize alternatives based on web surveys.Methodology/Approach: We use the response latency, i.e. the time taken to make a decision,to replace the rating scale in the pairwise comparisons which is a fundamental part of the AHPprocedure. A method combining the AHP with a mathematical model for the response latencyis proposed in the paper.Findings: Eliminating the rating scale and leaving only the “take the best” task alleviates therespondent and makes the survey much faster and intuitive. Moreover the results look morereliable in terms of consistency of judgments.Research Limitation/implication: An application to the development of a new service is usedto test the proposed method. Good results of the method are achieved also thanks to the use ofweb survey platforms, today easily accessible to both researchers and respondents; strengthsand weaknesses of web surveys are taken into consideration.Originality/Value of paper: The integration of the response latency in the traditional procedureof AHP gives two interesting contributions: on one hand it can be seen as a way to overcome,or at least alleviate, some limitations of the AHP; on the other hand it is a natural wayto involve the response latency, which is a measure quite well known in many scientific fields but seldom modeled and very rarely utilized in managerial decisions. Nowadays with the use ofweb surveys that measure can be much more easily manageable.

KW - Analytical Hierarchy Process; pairwise comparisons; response latency; relative importance
weights; prioritization

KW - web-survey

KW - Analytical Hierarchy Process; pairwise comparisons; response latency; relative importance
weights; prioritization

KW - web-survey

UR - http://hdl.handle.net/10447/95127

M3 - Other

SP - 178

EP - 191

ER -