PNNL team developed an interactive, dynamic and close-loop train-the-trainers application in the cybersecurity framework webtool. This application will have training scenarios that are designed based on major cyberattacks that surfaced in the United States (and potentially other countries) in the last decade. Through this training, the facility operators will go through an interactive game based hands-on training for the given scenarios. Under each scenario, every single cybersecurity aspect of the cybersecurity framework web tool will be weighted and allocated with resource value (represented in U.S. dollars). The application will use a closed loop constrained based optimization algorithm in the background to generate the resource value associated with a state change for each cybersecurity aspect. Prior to the beginning of the training, the facility operators could adjust the dollar values associated with the cybersecurity aspects to improve the accuracy or continue the training with the default values generated by the application. This will let the facility operators implement a comprehensive risk-based approach to identify cost-effective improvements to their facility to improve the cybersecurity posture. The output of such training may also be used to make investment decisions and establish related communications between operational technology (OT) and information technology (IT) teams. Therefore, the facility operators can perform a cost-benefit analysis and make economic decisions that reflect the improvement of the facility's cybersecurity posture. Current Stage of Development: A beta version of the game model is developed and is active. It is available at www.cybersecfw.org/game.stm. At this stage, the cyber scenarios aspect is in the development stage but the game is still playable using a test scenario (find the attachment in this form, load that into the web-tool and play). Eventually, PNNL will develop Easy, Medium, Hard versions (similar to the video games) based on the complexity of the cyber attack scenarios. Using Hill-climbing approach, the team identified the Gaussian distribution limits (min and max) per state and per level. At this stage, due to lack of cybersecurity data, the game is not based on any machine learning approach. This is developed to incorporate such extensions in the future. The dollar value associated with the questions are purely based on a mathematically optimized solution. For the intended purposes of this tool, the dollar value is to put the operator on a hard-decision-making spot rather than giving them a real dollar value to implement specific cybersecurity controls.