Abstract: Park-level integrated energy system (PIES) enables efficient utilization of energy and improves the consumption of renewable energy source. However, as the penetration of renewable energy source increases, the operational flexibility demand increases sharply, which poses great challenges to the safe and economic operation of the PIES. Consequently, this paper proposes an optimal dispatch method for the electric and thermal PIES, in which the risk of insufficient operational flexibility is taken into consideration. First, based on the comprehensive consideration of the operational flexibility demand of the electric system and non-electric (thermal) system, the mathematical expressions of the operational flexibility demand and supply of PIES are derived, and the operational flexibility margin and the risk cost of insufficient operational flexibility are taken as the evaluation indicators of the operational flexibility. Second, based on the conditional value-at-risk (CVaR), the mathematical expression of the risk soft constraint of insufficient operation flexibility is derived. Because the proposed optimal dispatch model of electric-thermal PIES considering the risk of insufficient operational flexibility is stochastic, non-convex, and nonlinear and has 0/1 integer variables, it belongs to NP-hard problem mathematically. Therefore, the proposed model is established under the framework of Markov decision process and is solved by approximate dynamic programming (ADP) algorithm. Finally, numerical simulations are carried out to verify the effectiveness of the proposed method.