I am doing a Discrete Choice Experiment (DCE) with surveys for my research. My research is based on skincare products with different attribute levels/types. With a total of four attributes (each with 3 levels/types), I have generated 81 different products. I have separated them into 3 different surveys (each with 9 choice sets, each choice set has 3 products to choose from and respondents can only choose one of the three). Moreover, they also answered questions about their demographic and familiarity with certain brands (brands I used in the product design). These are the covariates (age group, income group, region of residence, familiarity of use for 6 different brands).

I try to prepare dataframe for mixed multinomial logit model using mlogit package.
My data looks as follows:

I’m encountering an error when I try to use the fmlogit (fractional multinomial logit) package in R (https://github.com/f1kidd/fmlogit).

FYI: modelQ2_1 is a mixed logit model

Running the Mlogit function does not work, because it says that the two indexes(Resp_id & Alternative_id) are not unique. I have twelve times the same ID, because each respondent got 12 choice options with 3 alternatives.