We study linear policy approximations for the risk-

conscious operation of an industrial energy system with uncertain

wind power, significant and variable electricity demand, and

high thermal output, as found in a modern foundry. The

system incorporates thermal storage and operates under rolling

forecasts, leading to a sequential decision-making framework. To

address uncertainty in key parameters, we formulate chance-

constrained optimization problems that limit the probability

of critical constraint violations—such as unmet demand re-

quirements or the exceedance of system boundaries. To reduce

computational effort, we replace direct uncertainty handling

with a parameter-modified cost function that approximates the

underlying risk structure. We validate our method through

a numerical case study, demonstrating the trade-offs between

operational efficiency and reliability in a stochastic environment.