As the share of renewable sources in Europe's electricity mix continues to grow, manufacturers face both an opportunity and a challenge. Solar and wind generation introduce variability into the grid: electricity prices fluctuate throughout the day based on supply conditions, and the carbon intensity of available power shifts hour by hour. For energy-intensive production facilities, aligning operations with these fluctuations can yield significant cost savings and emissions reductions.
The SPEAR project addressed this opportunity directly through its optimisation algorithm library. Rather than treating electricity as a fixed-cost input, the platform's algorithms could factor in time-varying energy prices and grid conditions when recommending production schedules. This capability transformed energy from a passive overhead into an active variable that production planners could optimise alongside throughput and quality targets.
Load Shifting in Practice
The concept of load shifting is straightforward in principle: move energy-intensive operations to time windows where electricity is cheaper, cleaner, or both. In practice, however, implementation is far from simple. Production schedules are constrained by delivery deadlines, material availability, equipment maintenance windows, and quality requirements that may mandate specific process sequences or timing. Any load-shifting recommendation that ignores these constraints risks disrupting operations rather than improving them.
The SPEAR platform's advantage lay in its detailed knowledge of the energy consumption profile for every operation in the production sequence. Because the digital twin could predict exactly how much energy each process step required and when it required it, the optimisation algorithms could identify realistic shifting opportunities. A heat treatment step that consumed significant energy over 45 minutes might be schedulable in multiple time windows, while a precision machining operation with tighter constraints might need to remain fixed. The platform could reason about these trade-offs automatically.
Portuguese partners from GECAD at the Instituto Superior de Engenharia do Porto contributed substantial expertise in demand response and consumption management strategies. Their research within the project explored how industrial facilities could participate in formal demand response programmes, receiving financial incentives from grid operators in exchange for adjusting consumption patterns during peak demand periods.
The Textile Production Case
One of the project's demonstrators examined a textile production line that was partially powered by photovoltaic panels. The challenge was to schedule production operations so that maximum use was made of on-site solar generation, reducing both grid electricity costs and the facility's carbon footprint. A genetic algorithm was developed to explore the vast space of possible production schedules and identify configurations that maximised self-consumption of solar energy while meeting all production deadlines.
The results showed that intelligent scheduling could meaningfully increase the proportion of production powered by on-site renewable generation, without requiring any additional investment in solar capacity. The improvement came purely from aligning the timing of energy-intensive operations with periods of high solar output. This type of "free" efficiency gain, requiring no capital expenditure beyond the optimisation tool itself, is particularly attractive for small and medium-sized manufacturers operating on tight margins.
Implications for Grid Stability
From a wider perspective, industrial load shifting also benefits electricity grid stability. As renewable penetration increases, grid operators face growing challenges in balancing supply and demand. Manufacturing facilities that can flexibly adjust their consumption patterns help absorb surplus renewable generation during periods of high output and reduce stress on the grid during periods of scarcity.
The SPEAR platform's ability to forecast energy consumption at the component level made it possible to offer grid operators reliable flexibility commitments. Unlike simple on/off demand response, where a facility either curtails all production or none, the granular approach allowed selective adjustment of specific processes while maintaining overall production continuity. This precision made industrial demand flexibility a more attractive proposition for both manufacturers and grid operators.
As European energy markets continue to evolve with increasing renewable integration and dynamic pricing structures, the ability to optimise production schedules against real-time energy conditions will become a significant competitive factor for manufacturing companies of all sizes.