To improve the global performance of the standard teaching–learning-based optimization (TLBO) algorithm,
an improved TLBO algorithm (LETLBO) with learning experience of other learners is proposed in
the paper. In LETLBO, two random possibilities are used to determine the learning methods of learners
in different phases. In the Teacher Phase, the learners improve their grades by utilizing the mean information
of the class and the learning experience of other learners according to a random probability. In
Learner Phase, the learner learns knowledge from another learner which is randomly selected from the
whole class or the mutual learning experience of two randomly selected learners according to a random
probability. Moreover, area copying operator which is used in Producer–Scrounger model is used for parts
of learners to increase its learning speed. The feasibility and effectiveness of the proposed algorithm are
tested on 18 benchmark functions and two practical optimization problems. The merits of the improved
method are compared with those of some other evolutionary algorithms (EAs), the results show that the
proposed algorithm is an effective method for global optimization problems.