The invention relates to a modeling method and device for automatically organizing key performance indicators. According to the method, based on an object
element model and a relation meta model of
power grid running KPI data, the expansion analysis technology for
data mining is utilized for setting up an indicator
pool, then the KPI data in the indicator
pool are automatically organized according to the time dimension, the space dimension and the service dimension to set up a three-dimensional coordinate
system, on the basis of the three-dimensional coordinate
system, the KPI data are automatically organized secondarily according to the
process dimension, the object dimension and the target dimension to construct a KPI cube to form a six-dimensional automatic key performance indicator organizing model, and then according to a decision-making objectives of
power grid service, the KPI cube is segmented along the
process dimension, the object dimension and the target dimension for automatically organizing a key performance indicator set. According to the method, according to different kinds of service and the decision-making objectives, the key performance indicators and correlation control strategies can be organized automatically and efficiently, and by setting up the automatic key performance indicator organizing model, the actual running state of a
power grid can be mastered fast, accurately and comprehensively.