Depends why you think a person did not use it.

For a factor analysis, I would treat a "did not use" score as a missing score. It does not make sense to score the "did not use" category on the same dimension as the satisfaction categories.

Not all the available services are to be used by everyone. Thus felt the need for extra option as "Did not use". As in, money changers/ATM or parking facilities if the passengers use taxi, are not used by everyone. So, they can't rate those services. However, others may use the service and need to rate those facilities.

What are you trying to do with the factor analysis? It might not make sense to use it.

Could try optimal scaling with several options. Treat missing as a new category for everyone or certain groups of respondent, but could also scale as a new category for every single person. I know this option are available in the Categories module from SPSS, but not sure if this has been implemented in packages at R.

Diwash Malla Thakuri, you raise an important issue that relates to collecting valid data from an appropriate sample. Too often researchers don't provide a "Did not use" or equivalent option (e.g., "Not applicable"), so participants resort to using other options that have indeterminate meanings (e.g., an "Unsure" option if it's provided), or they simply stop responding to a questionnaire because they become frustrated at the lack of appropriate response-option offerings.

In relation to your question about what's best to do in factor analysis (I assume you are referring to exploratory factor analysis), particularly if only a few participants have used the "did not use" option that was given them, I think it's best to convert the "Did not use" responses to missing, as suggested by Christian Geiser up above. Then I recommend you conduct EFA by selecting "Exclude cases pairwise" under the Options tab. That will mean you lose only part of the data from some participants rather than losing all of the data from participants who responded to any of the items with "Did not use".

I hope that makes sense. Do feel free to get back if you have any queries.

All the best for your research.

As Robert Trevethan notes, it is good that you included a choice for "Did not use." The best treatment of the data depends on what question you are trying to answer. If you are looking at the satisfaction of users, then the opinions of non-users are generally not meaningful, making it generally proper to omit the data (though there may be a possibility that someone was a non-user because of problems with the service; e.g., if I don't check my luggage, it may be because I don't trust them with my luggage). If you are looking at the usefulness of a service, then non-use of the service is meaningful data.