Understanding business drivers and improving business forecast
Factor analysis
The purpose of factor analysis is to simplify a rather complex collection of facts and
figures by converting the variables into unobservable variables called factors. In 1997
Toyota Motors Corporation used factor analysis to help them analyze whether an invention
was good or bad. An interview was conducted in which the engineers were asked to define
what they thought was a good invention. The results of the interview and factor analysis
were used to analyze and decide the factors for categorizing a product as being good.
Multidimensional scaling
Multidimensional scaling is a method in which a visual representation of data is charted up
to show the relationship between two sets of data. Multidimensional scaling is an excellent
way of comparing two different products in the market. t usually on the demand of a
customer. The Miller Fillmore Hospital in New York used multidimensional scaling to
compare MFH with other hospitals in New York to see how the MFH fared against other
hospitals and also to find out the what their strategy lacked. The desired outcome of these
studies was to increase the number of patient referrals and increase the number of
physicians.
Cluster analysis
Cluster analysis is a statistical method used to group together certain types of data together
that share the same attributes. On September 4th 1999 researchers at Sandia labs
successfully used cluster analysis to determine the types of explosions taking place in New
Mexico, Wyoming. The use of cluster analysis before this research was usually limited to
the field of medicine and biology. With the help of cluster analysis the researchers were
able to tell the nature of the seismic activity.