Enhancing Industrial Flexibility and Market Participation in Cement Manufacturing Through Optimized Production Scheduling

The growing share of variable renewable energy (VRE) sources in power systems is increasing the need for short term operational flexibility, particularly from large industrial electricity consumers. This study proposes a practical, two stage optimization framework to unlock this flexibility in cement manufacturing and support participation in electricity balancing markets. In Stage 1, a mixed integer linear programming (MILP) model minimizes electricity procurement costs by optimally scheduling the raw milling subsystem. In Stage 2, a flexibility assessment model evaluates profitable deviations, targeting participation in Spain manual Frequency Restoration Reserve (mFRR) market. A real world case study in a Spanish cement plant (including PV and battery storage) shows that flexibility services can yield monthly revenues of up to 800 EUR and paybacks as short as six years. This framework offers a replicable pathway for industrial flexibility in energy intensive sectors.

💡 Research Summary

The paper addresses the growing need for short‑term operational flexibility from large industrial electricity consumers as variable renewable energy (VRE) penetration rises. Focusing on the cement industry—responsible for roughly 7‑8 % of global CO₂ emissions and about 7 % of industrial energy use—the authors target the raw milling subsystem, which accounts for 20‑30 % of a plant’s electricity demand but is decoupled from the continuous kiln operation via an intermediate silo. This structural decoupling provides inherent flexibility that can be monetized in electricity balancing markets.

A two‑stage mixed‑integer linear programming (MILP) framework is proposed. Stage 1 determines a cost‑optimal production schedule for the raw mill using day‑ahead electricity prices, PV generation forecasts, and battery storage (if present). Decision variables include binary on/off status of each mill, silo inventory levels, grid import/export power, and battery charge/discharge power. Constraints enforce material balance in the silo, power balance among grid, PV, battery, and mills, inventory limits, kiln demand satisfaction, minimum on/off times, and battery state‑of‑charge dynamics with depth‑of‑discharge limits. The objective minimizes total electricity procurement cost (grid purchases only).

Stage 2 builds on the baseline schedule to evaluate participation in Spain’s manual Frequency Restoration Reserve (mFRR) market, a tertiary regulation service with up‑to‑15 minute lead time and up to two‑hour duration. The model quantifies the cost of deviating from the baseline (e.g., offering ±ΔP MW at time τ) while respecting all operational constraints, and computes the net economic gain by subtracting the additional electricity cost from the market remuneration received for the offered reserve.

The methodology is validated on a real Portland cement plant in Spain. Using 2023 market data, the authors examine 19 hypothetical configurations of on‑site photovoltaic (PV) generation (1‑6 MW) and battery energy storage systems (BESS) (1‑6 MWh). Simulations are run over rolling horizons of 24 to 168 hours with hourly resolution. Key findings include:

• PV self‑consumption and battery dispatch reduce electricity procurement costs by 3‑5 % on average.
• Participation in the mFRR market yields additional monthly revenues of €400‑€800, with a marked increase when battery capacity exceeds 4 MWh.
• Simple payback periods for the BESS investment range from 5 to 7 years; the most favorable configurations achieve payback in under six years.

Technical contributions are: (i) a detailed MILP formulation that captures raw mill binary operation, silo inventory dynamics, and battery cycling costs; (ii) integration of real market price signals into both cost‑minimization and flexibility‑valuation stages; (iii) a comprehensive techno‑economic assessment across multiple renewable‑storage scenarios, providing actionable guidance for industrial stakeholders.

The study fills a gap in the literature where prior works either focused on continuous processes (e.g., steelmaking) or considered only a single market layer. By jointly optimizing production scheduling and market‑responsive flexibility for a non‑continuous, inventory‑buffered process, the paper demonstrates a replicable pathway for energy‑intensive industries to monetize demand‑side flexibility without disrupting core production.

Limitations include the deterministic treatment of electricity prices and PV output, and the omission of equipment failure or maintenance schedules, which could affect real‑world feasibility. Future research directions suggested are stochastic scenario‑based optimization, multi‑market participation (primary, secondary, tertiary, and ancillary services), and integration with real‑time control platforms for on‑line dispatch.