You may like to examine the product X_i*I-i as the new index of performance that incorporates CSI as well as volume, and use this for ranking. This may or may not serve the purpose. ~Abhaya -- *Dr Abhaya Indrayan * *Medicalizing Biostatistics: http://www.MedicalBiostatistics.com<http://www.medicalbiostatistics.com/> * *Personal website: http://indrayan.weebly.com Encyclopedic book: http://MedicalBiostatistics.synthasite.com<http://medicalbiostatistics.synthasite.com/> * * * On Mon, Aug 1, 2011 at 11:10 PM, Sushmita <[log in to unmask]> wrote: > Dear Ali, > You can assign some weights to 'reception' and 'CSI', based on their > importance. For e.g take weights w_1 and w_2 such that w_1 + w_2 = 1. If CSI > is of more importance then one could assign w_1 = .2 to reception and w_2 = > 0.8 to CSI and one can calculate w_1 * reception + w_2 * CSI for every > repair shop and select accordingly. > I hope this works for you. > All the best! > Sushmita > --- On Mon, 8/1/11, Ali Akbar Bromideh <[log in to unmask]> wrote: > > From: Ali Akbar Bromideh <[log in to unmask]> > Subject: Query on Ranking. > To: [log in to unmask] > Date: Monday, August 1, 2011, 4:22 PM > > Dear Sir/Madam, > > I have been asked to submit this question to the list. Consider a company > has n after sales outlets (repair shops), namely X_1, X_2, ..., X_n. Monthly > reception by each repair shop are recorded as V_1, V_2, ... V_n and their > monthly customer satisfaction index (CSI) are I_1, I_2, ..., I_n. Now the > question is how to rank the repair shops? > > For an instance, consider the simplified case: > > Repair shops reception(unit/month) CSI (%) > X_1 100 51 > X_2 20 87 > X_3 50 60 > X_4 200 63 > > Note that ignoring the reception volume, the ranks are: X_2, X_4, X_3, X_1. > What are the ranks of shops considering the monthly receptions? > > Kind regards, > Ali > > You may leave the list at any time by sending the command > > SIGNOFF allstat > > to [log in to unmask], leaving the subject line blank. > > You may leave the list at any time by sending the command > > SIGNOFF allstat > > to [log in to unmask], leaving the subject line blank. > You may leave the list at any time by sending the command SIGNOFF allstat to [log in to unmask], leaving the subject line blank.