@9:19 under "communication skills" (Candidate 4) shouldn't the numerator of the c component be 5 instead of 9? And under "knowledge" (Candidate 4) numerator of a,b and c component is entirely different....
Hello! I'm writing this comment to help people that are trying to learn about fuzzy TOPSIS. These videos are generally a good introduction to the topic and im thankful for them. They're certainly easier to follow then formulas from articles. However, in this particular video there's a pretty serious mistake. While determining ideal solutions you should: 1) In the case of FPIS Determine a single maximum number from the c factors in a column. For criteria nr.1 from the video it would be 9. Construct a vector (fuzzy number) consisting ONLY of the said number. For criteria nr. 1 from the video the vector would be (9,9,9) 2) In the case of FNIS Do the same, except with minimum number from the a factors in a column. For criteria nr. 1 from the video the vector would be (0,556; 0,556; 0,556) Also, there are 2 more ways to determine FPIS and FNIS: 1) For FPIS use a vector of 1s (1,1,1) and for FNIS use a vector of 0s (0,0,0) (for every criteria/column). 2) For FPIS use maximum values and for FNIS use minimum values. For criteria nr. 1 from the video the FPIS = (2,778; 6,481; 9) and FNIS = (0,556; 1,815; 5). In other words, for FPIS use the greatest and for FNIS smallest a, b, and c factors. I hope this will be helpfull. Cheers. For more information, please, head to this article: "Evaluating R&D Projects as Investments by Using an Overall Ranking From Four New Fuzzy Similarity Measure-Based TOPSIS Variants" by Mikael Collan and Pasi Luukka (2014)
The publication of Chen (2000) Extensions of the TOPSIS for group decision-making under fuzzy environment, the aggregation os preference of each decision maker in made by the mean of each xij, and not a function of min, mean and max as you did.
Agree, I would also like to see a publication with weighted aggregation of the preferences of each decision maker (in cases where some decision-makers are more expert than others, so this should be represented in the aggregation as well)
Hi, how can I get the weight for each criteria? Can you explain please? I tried with AHP fuzzy weight but i cant understand how in this example you get HIGH : 5 7 9 and so on… Thanks for your helping
You are a great man. Thanks for video. It will be good on coronadays. I couldnt go to school, but I have to write my article. It is so benefical for me. Thank you again.
Hello. Thank you very much for the insightful video. I have one concern, though. When there are multiple decision makers in weighing the criteria when using FAHP, can we apply combined decision matrix as well or we need to apply another method?
Hello sir, Pls cite the source from where you have taken formula to combine all three decision matrix. Also, this same formula can be used to combine the pairwise matrix of decision makers in case of fuzzy ahp? Pls help sir.
Tell me, please, why are you considering the cost criteria "Time to solve a problem" ? Can we apply this method without cost criteria, just having all benefits criteria? Thank you and congratulations for the videos.
Hi sir, may I know the basis for the weightage selection in the combined decision matrix? why are high, very high, and average considered in their respective columns?
If you are giving different weightage to different decision maker based on the experience or hierarchy. those weights will be multiplied in their decision value. That will give you a weighted average value when calculating a combined matrix.
Very Helpful. An excellent way of explaining the complete process. Can you please share the article/literature for citation of the formulas and process. Thanks
Thanks for the complement, I have updated my wordpress site. In the bottom of the page there is a reference section in which i have cited papers which i have used for making videos.
If you have a mixture of qualitative and quantitative values of criteria, then all the qualitative values will be converted to quantitative value using a fuzzy 7 point scale or fuzzy 11 point scale.
@@faizabouchraki5580 i will send u the paper which have a 11 point fuzzy scale example. mail me at mathewmanojraipur@gmail.com so that i can revert back the attachment.
@@manojmathew5287 is it compulsory to use fuzzy five scale for qualitative data and fuzzy 7 and 9 for both qualitative and quantitative data? your lecture helps me alot just try to make video on generating fuzzy numbers
Thanks for sharing. I am confused about the benefit criteria and cost criteria. These criteria are taken into consideration? or some process to decide it? Fuzzy topsis can apply for dependant variables like Surface finish, flank wear in manufacturing problem?
yes for both beneficial and non beneficial, max. value is taken for FPIS. The values are normalised at the initial stage so automatically the best value will take the highest value.
Thank you so much Manoj and kudos to you for making auch informative videos and explaining such mathematical intensive topics with such clarity. Cheers.
Hey, thank you for your helpful videos! I would like to solve the phone problem (like in fuzzy AHP) using Fuzzy TOPSIS method but only with 1 decision maker (individual), is it similar to what you showed in video from some point or completely different? I would be really grateful for some clues. Greetings
Hi Mr. Mathew, does the fuzzy numbers assigned for the linguistic variables have to be this way or can it vary? for example in Fuzzy AHP very low=(1,1,1) while Fuzzy TOPSIS has (1,1,3) is assigned to very low? Thanks
Hi thank you for such great video! I want to ask: if there is only one candidate but multiple decision makers, this method can be applied to classify the opnions about the candidate?
There are some errors in the calculations presented in this video. 1. The weight values should be normalized if their sum is more than 1. This is the case of the weight values in this video, but Mr. Manoj Mathew has not normalized them. 2. The second error concerns the values of positive and negative optimal solutions, which are A* and A-, respectively. Time in the example presented here is the cost value and its minimal value should be taken as A* and its maximal value should be taken as A-, whereas the inverse cases have been considered here. In other words, in this video the maximal value and the minimal value for time have been taken as A* and A-, respectively. This is the case for beneficial criteria, whereas time is a cost criterion and this consideration for its A* and A- is not correct. So, the procedure and the output in this video are false. Mr. Mathew should correct these errors in this video and apologize for having published them on the Web since 3 years ago. As he is not familiar with TOPSIS, he has to learn it correctly before publishing anything on this issue on the Web in order to not causing mistakes in his visitors' calculations.
Thank you for being a critic. I have followed all calculations from the below given paper. 1) In normal TOPSIS you normalize if their sum is more than 1, but in fuzzy TOPSIS this is not done in the same way as in simple TOPSIS because we use fuzzy numbers instead of simple crisp number. Still we have followed the normalization (refer 7.04 of the video) 2) Please refer step-5 of the paper (link given below), there is no mention of cost or beneficial criteria in the computation of FPIS and FNIS, as we have already considered the cost or beneficial criteria in the normalization step, so no need of doing it in the computation of FPIS and FNIS. (If you compare normal TOPSIS with fuzzy TOPSIS then there are few changes: Vector normalization is done in simple TOPSIS where we don't consider cost or beneficial criteria, but in case of fuzzy TOPSIS we consider cost or beneficial criteria in the normalization step) Hope your doubts are clear. Please refer this paper for more details: Nădăban, S., Dzitac, S., & Dzitac, I. (2016). Fuzzy TOPSIS: a general view. Procedia computer science, 91, 823-831.
Hi Manoj, Can you please let me the difference between different Multi-criteria Decision matrics (MCDM)? and how to chose the best one? I want to use MCDM for risk assessment in construction but don't wana use AHP and ANP can you let me know the benefits of TOPSIS, ELECTRE over AHP or ANP?
Sir at 11:50 for time to solve problem which is cost criteria you have taken max value for A* while in other video you have considered min value for cost criteria for A*. How is that different from non-cost criteria considering max value for both cost and non-cost criteria ?
yes for both cost and non cost criteria, max. value is taken for FPIS calculation. The values are normalised at the initial stage so automatically the best value will take the highest value.
@@manojmathew5287 Sir, do we have standard fuzzy conversion tables available for each of the different membership functions like triangular of trapezoid. If so could you please post the link for those standard tables here. ?
iam sorry Mr. Manoj. in document Fuzzy.topsis_calculation, it explains the weights of only 3 variables, namely; high, very high, and average. whereas in Fuzzy Number there are low and very low. explanation please
i would to thank you for this work who is really relevant.i would to know the difference between the benefit and cost criteria. and how we can identify them.thank you very much
benefit criteria means the variables whose higher values are desired, e.g. skill, whereas cost criteria means variables whose lower values are desired,e.g. price of mobile and time taken by some employee to complete a task.
@9:19 under "communication skills" (Candidate 4) shouldn't the numerator of the c component be 5 instead of 9? And under "knowledge" (Candidate 4) numerator of a,b and c component is entirely different....
Ya thats a typographical error. further calculations are correct, used the actual value
Hello! I'm writing this comment to help people that are trying to learn about fuzzy TOPSIS. These videos are generally a good introduction to the topic and im thankful for them. They're certainly easier to follow then formulas from articles. However, in this particular video there's a pretty serious mistake. While determining ideal solutions you should:
1) In the case of FPIS
Determine a single maximum number from the c factors in a column. For criteria nr.1 from the video it would be 9.
Construct a vector (fuzzy number) consisting ONLY of the said number. For criteria nr. 1 from the video the vector would be (9,9,9)
2) In the case of FNIS
Do the same, except with minimum number from the a factors in a column. For criteria nr. 1 from the video the vector would be (0,556; 0,556; 0,556)
Also, there are 2 more ways to determine FPIS and FNIS:
1) For FPIS use a vector of 1s (1,1,1) and for FNIS use a vector of 0s (0,0,0) (for every criteria/column).
2) For FPIS use maximum values and for FNIS use minimum values. For criteria nr. 1 from the video the FPIS = (2,778; 6,481; 9) and FNIS = (0,556; 1,815; 5). In other words, for FPIS use the greatest and for FNIS smallest a, b, and c factors.
I hope this will be helpfull. Cheers. For more information, please, head to this article: "Evaluating R&D Projects as Investments by Using an Overall Ranking From Four New Fuzzy Similarity Measure-Based TOPSIS Variants" by Mikael Collan and Pasi Luukka (2014)
The publication of Chen (2000) Extensions of the TOPSIS for group decision-making under fuzzy environment, the aggregation os preference of each decision maker in made by the mean of each xij, and not a function of min, mean and max as you did.
Agree, I would also like to see a publication with weighted aggregation of the preferences of each decision maker (in cases where some decision-makers are more expert than others, so this should be represented in the aggregation as well)
very helpful video....amazing one...thank you Dr Manoj...thanks a lot...you have really done a tremendous and amazing work.
Thanks a lot for the videos! Helps me a lot with my theses.
Hi, how can I get the weight for each criteria? Can you explain please? I tried with AHP fuzzy weight but i cant understand how in this example you get HIGH : 5 7 9 and so on… Thanks for your helping
Thanks for the amazing work, please post videos on grey method
how we can change the Likert Scale to fuzzy numbers?
Thank you for the wonderful explanation.
Very helpful material to understand this method.
Thanks for the explanations. It would be nice to see the performance of Fuzzy PROMETHEE in your videos.
Thank you so much for the videos! ✅
You are a great man.
Thanks for video. It will be good on coronadays.
I couldnt go to school, but I have to write my article. It is so benefical for me.
Thank you again.
Hello. Thank you very much for the insightful video. I have one concern, though. When there are multiple decision makers in weighing the criteria when using FAHP, can we apply combined decision matrix as well or we need to apply another method?
Can I use the fuzzy numbers of Fuzzy TOPSIS with Fuzzy SAW??
Hello sir, Pls cite the source from where you have taken formula to combine all three decision matrix.
Also, this same formula can be used to combine the pairwise matrix of decision makers in case of fuzzy ahp?
Pls help sir.
Sir, PLEASE make a video on implementation of fuzzy topsis on matlab.
Tell me, please, why are you considering the cost criteria "Time to solve a problem" ? Can we apply this method without cost criteria, just having all benefits criteria? Thank you and congratulations for the videos.
Hi sir, may I know the basis for the weightage selection in the combined decision matrix? why are high, very high, and average considered in their respective columns?
how do i understand beneficial and cost criteria? please enlighten me.
How you assigned it is for beneficial criteria or for non beneficial criteria ???
How to proceed if we give different importance to different decision makers?
If you are giving different weightage to different decision maker based on the experience or hierarchy. those weights will be multiplied in their decision value. That will give you a weighted average value when calculating a combined matrix.
How to create the decision maker
Very Helpful. An excellent way of explaining the complete process. Can you please share the article/literature for citation of the formulas and process. Thanks
Thanks for the complement, I have updated my wordpress site. In the bottom of the page there is a reference section in which i have cited papers which i have used for making videos.
Great content, thank you!
can i use a pairwise comparison for fuzzy AHP as the first step of Fuzzy topsis??
How AHP and TOPSIS work together?? can you please guide me
Plz Dr,if i have one decision maker, i don't need to do a step of combined decision matrix????
Sir name of this method HYBRID FUZZY AHP AND TOPSIS ....?
if we have a mixed of qualitative and quantitative criteria and we want to use a fuzzy logic how we do it .thank you very much
If you have a mixture of qualitative and quantitative values of criteria, then all the qualitative values will be converted to quantitative value using a fuzzy 7 point scale or fuzzy 11 point scale.
@@manojmathew5287 in this case so we don't use triangular membership function ?? you can give me an example about the fuzzy 7 point scale please ???
@@faizabouchraki5580 i will send u the paper which have a 11 point fuzzy scale example. mail me at mathewmanojraipur@gmail.com so that i can revert back the attachment.
@@manojmathew5287 is it compulsory to use fuzzy five scale for qualitative data and fuzzy 7 and 9 for both qualitative and quantitative data? your lecture helps me alot just try to make video on generating fuzzy numbers
Thanks a lot and i have i question
how we can determine if a criteria is benefit or cost?
Thank you so much 🙏🏽
if using pythagorean fuzzy with with only two scale?
Thanks for sharing. I am confused about the benefit criteria and cost criteria. These criteria are taken into consideration? or some process to decide it? Fuzzy topsis can apply for dependant variables like Surface finish, flank wear in manufacturing problem?
Hi @11:55 , are the FPIS values the max. Vij values for both the beneficial and non-beneficial criteria ( time to solve the problem) ?
yes for both beneficial and non beneficial, max. value is taken for FPIS. The values are normalised at the initial stage so automatically the best value will take the highest value.
Thank you so much Manoj and kudos to you for making auch informative videos and explaining such mathematical intensive topics with such clarity. Cheers.
reference please!
do you have matlab code for fuzzy topsis
Hey, thank you for your helpful videos! I would like to solve the phone problem (like in fuzzy AHP) using Fuzzy TOPSIS method but only with 1 decision maker (individual), is it similar to what you showed in video from some point or completely different? I would be really grateful for some clues.
Greetings
Its same for single decision maker
Hi Mr. Mathew, does the fuzzy numbers assigned for the linguistic variables have to be this way or can it vary? for example in Fuzzy AHP very low=(1,1,1) while Fuzzy TOPSIS has (1,1,3) is assigned to very low? Thanks
Fuzzy numbers defined for linguistic variables are decided by decision makers, it can vary person to person
@@manojmathew5287 Sir, Can you please explain, how did you assign the weight?
can we use fuzzy topsis if number of alternatives are more than 1 million? please reply
Hi thank you for such great video! I want to ask: if there is only one candidate but multiple decision makers, this method can be applied to classify the opnions about the candidate?
There are some errors in the calculations presented in this video. 1. The weight values should be normalized if their sum is more than 1. This is the case of the weight values in this video, but Mr. Manoj Mathew has not normalized them. 2. The second error concerns the values of positive and negative optimal solutions, which are A* and A-, respectively. Time in the example presented here is the cost value and its minimal value should be taken as A* and its maximal value should be taken as A-, whereas the inverse cases have been considered here. In other words, in this video the maximal value and the minimal value for time have been taken as A* and A-, respectively. This is the case for beneficial criteria, whereas time is a cost criterion and this consideration for its A* and A- is not correct. So, the procedure and the output in this video are false. Mr. Mathew should correct these errors in this video and apologize for having published them on the Web since 3 years ago. As he is not familiar with TOPSIS, he has to learn it correctly before publishing anything on this issue on the Web in order to not causing mistakes in his visitors' calculations.
Thank you for being a critic.
I have followed all calculations from the below given paper.
1) In normal TOPSIS you normalize if their sum is more than 1, but in fuzzy TOPSIS this is not done in the same way as in simple TOPSIS because we use fuzzy numbers instead of simple crisp number. Still we have followed the normalization (refer 7.04 of the video)
2) Please refer step-5 of the paper (link given below), there is no mention of cost or beneficial criteria in the computation of FPIS and FNIS, as we have already considered the cost or beneficial criteria in the normalization step, so no need of doing it in the computation of FPIS and FNIS. (If you compare normal TOPSIS with fuzzy TOPSIS then there are few changes: Vector normalization is done in simple TOPSIS where we don't consider cost or beneficial criteria, but in case of fuzzy TOPSIS we consider cost or beneficial criteria in the normalization step)
Hope your doubts are clear.
Please refer this paper for more details:
Nădăban, S., Dzitac, S., & Dzitac, I. (2016). Fuzzy TOPSIS: a general view. Procedia computer science, 91, 823-831.
Thanks for the explanations
hi sir i would like to know if there are any methods for plannification and scheduling ???
Hi Manoj, Can you please let me the difference between different Multi-criteria Decision matrics (MCDM)? and how to chose the best one? I want to use MCDM for risk assessment in construction but don't wana use AHP and ANP can you let me know the benefits of TOPSIS, ELECTRE over AHP or ANP?
Thanks sir, when the criteria in numerical form like engineering output then how to use fuzzy topsis?
Sir could you please perform sensitivity on same example of fuzzy topsis
Sir at 11:50 for time to solve problem which is cost criteria you have taken max value for A* while in other video you have considered min value for cost criteria for A*. How is that different from non-cost criteria considering max value for both cost and non-cost criteria ?
yes for both cost and non cost criteria, max. value is taken for FPIS calculation. The values are normalised at the initial stage so automatically the best value will take the highest value.
thanks for the amazing content
can you make a video on Fuzzy Gamma technique
Can u mail me the research paper which have fuzzy gamma technique so that i may have a look at the algorithm. mathewmanojraipur@gmail.com
Thank you sir for your valuable video. I have one question, how i generate the fuzzy number?
fuzzy numbers can be generated using fuzzy conversation scale/table
@@manojmathew5287 Sir, do we have standard fuzzy conversion tables available for each of the different membership functions like triangular of trapezoid. If so could you please post the link for those standard tables here. ?
sir how did you assign fuzzy numbers to the lingusistic terms? is it random or there is a pattern?
iam sorry Mr. Manoj. in document Fuzzy.topsis_calculation, it explains the weights of only 3 variables, namely; high, very high, and average. whereas in Fuzzy Number there are low and very low. explanation please
Tq so much
excellent
sir i think there is an error on 10:36 , (row4 ,column 3) : cells(4,3) : 3 *0.33 ; 5*1 ,1*7 ,you should multiply 5 not 3
yes thats a typo mistake
i would to thank you for this work who is really relevant.i would to know the difference between the benefit and cost criteria. and how we can identify them.thank you very much
benefit criteria means the variables whose higher values are desired, e.g. skill, whereas cost criteria means variables whose lower values are desired,e.g. price of mobile and time taken by some employee to complete a task.
12 3 0.934
12 8 0.292
6 8 0.6
8 3 1
8 1 1
16 2 0.8
this my alternatives , how to convert these number to fuzzy ???
Sir! PLs try to make a video on Fuzzy ELECTRE method thank you in advance
Thanks for the explanations