Industrial heat above moderate temperatures is hard to decarbonise due to constraints of energy carriers with regard to temperature quality, operational continuity, process integration and economic costs. We build an hourly sector-integrated optimization model for an industrial cluster in Kazakhstan that comprises the production of steel, cement, ammonia, and equivalent chemicals in terms of energy requirement. Our analysis accounts for low-, medium- and high-temperature heat requirements and integrates the procurement of renewable electricity, direct electricity heating, electrolysis, hydrogen storage, hydrogen combustion, heat recovery, and thermal storage. We compare four configurations of energy systems: one based on fossil fuels, one relying on direct electrification only, a mixed strategy involving hydrogen and electricity, and finally a fully sector-integrated configuration featuring cascading of waste-heat and multi-level thermal storage. The fully sector-integrated system leads to a drop in annual CO2 emissions from 5.94 to 0.19~Mt/year (96.8% reduction), and to a 16.5% decline in costs compared to the fossil baseline, and 8.7% compared to the direct electrification scenario. Hydrogen is reserved for reductant, feedstock and high-temperature heat, with recovered heat covering low-temperature heat service and electricity supplying medium-temperature heat service in the optimal solution. Thermal storage helps increase the effective usage of waste heat from 39% to 71%, and decreases the hydrogen consumption for combustion by 18.4% in comparison to the hybrid case. The results show that Kazakhstan’s industrial-heat transition is most credible when designed as a cluster-level thermodynamic integration problem rather than as isolated plant-level fuel substitution.