Why does grid congestion increasingly determine when a cold store can open?

In April 2026, Netbeheer Nederland reported around 15,000 pending connection requests with the regional grid operators, together representing 9,305 MW. A year earlier there were 12,000, representing 6,739 MW. The queue is not only getting longer; the power being requested is rising with it.

The sector itself draws the same conclusion. Nekovri, the Dutch association of cold and freezer stores, is campaigning to have the cold chain formally recognised as critical infrastructure, pointing to grid congestion and high energy tariffs eroding its members’ capacity to invest and compete: operators want to expand and decarbonise, and cannot get onto the grid.

That does not prove grid capacity governs every cold-store project. It does mean “capacity will be there in time” is an assumption to be tested during feasibility, rather than a given to build a programme on.

At the Rotterdam Food Hub in Europoort, the Port of Rotterdam Authority, Necron Group and PTP Group are jointly developing 38 hectares. Necron is building not only two cold and freezer stores totalling roughly 145,000 m², but also an energy hub on a third plot, intended to supply both the terminal and the cold stores from the start. Final investment decisions are expected in early 2027.

Another route is to connect to capacity that is already available nearby. Developers and operators may explore whether an existing connection held by a neighbouring business, a direct line or a shared energy system can provide access to sufficient power. This means that waiting for a new or upgraded public-grid connection is not necessarily the only route to commissioning.

These solutions are not interchangeable. An area-wide energy hub serving several parties, a direct connection between two neighbours and a closed distribution system are different constructions, with different owners, permitting requirements and risks.

How much power you need follows from choices that sit inside the warehouse design: the refrigeration concept and design conditions, building dimensions and insulation, the number of cranes, conveyors and lifts, the charging strategy for MHE and trucks, redundancy in refrigeration plant and transformers, and the required start-up and recovery behaviour. Conversely, those choices cannot be frozen without knowing what transport capacity is available or can be built locally.

So “bring energy to the table earlier” understates it. Power supply and operating architecture are the same design problem from feasibility onwards. Handle them in sequence and the feedback loop only appears once one of the two is fixed — at which point the correction is no longer a technical adjustment but an operating-model redesign.

The same applies to flexibility. A cold store has a non-negotiable cooling obligation, but that does not mean every electrical load is fixed at every moment: the thermal mass of the building and the stored product makes part of the cooling load temporarily shiftable within temperature and food-safety limits. How much room there is depends on the temperature regime — frozen stores are already being used as energy buffers, while chilled product leaves less margin.

This is not theoretical. Cold and freezer stores take part in congestion management by temporarily turning down cooling and freezing capacity, and a capacity-limitation contract formalises exactly that: accepting payment in return for not using the full contracted transport capacity. Feasibility should therefore establish which loads can be shifted or shed, for how long, and what recovery peak follows.

An energy hub, direct line or closed electricity system moves part of the responsibility from the public grid operator to the companies involved. Who operates the installation? Who is accountable for availability and continuity of supply? What covers maintenance windows? And when demand peaks, who decides which load takes priority?

That shift is also legal. In the Netherlands only recognised operators may run an energy network, and according to RVO grid congestion on its own is not grounds for such recognition — other reasons, such as safety and operational reliability on site, have to apply as well. A private network is not a technical shortcut around the waiting list.

The consequence is operational. Recovery from a power-supply failure becomes a process you design, test and rehearse, just like recovery from a WMS, WCS or MHE failure.

Leadership does not need to select the battery, CHP unit or contract structure itself. It does need to set three principles explicitly: what base and peak power the real operation requires, how much flexibility the process can tolerate, and what happens when the main supply route or a critical component fails.

It is worth being precise about what a solution actually does. A battery can flatten peaks or bridge short interruptions, but it does not automatically replace sustained backup supply; the question is always how many MW it delivers, how many MWh are available and which loads stay connected.

Only once those principles are fixed do the construction programme and automation choices become credible. Otherwise a tidy project plan gets mistaken for a site that can run. We have seen this before in logistics implementations: the building is ready, the systems are tested, everyone is lined up. The one condition carrying the whole operation is still an assumption.