Iberian Container Glass: Olive Oil, Wine, and the Export Squeeze

In the first half of 2023, EVOO fill volumes at Spanish container glass plants fell by an estimated 35-40%. Not a scheduling problem. Spain's 2022-23 olive harvest had collapsed to approximately 663,000 tonnes, less than half the five-year average of 1.4 million tonnes, and IS machines across Andalusia and Catalonia felt the under-demand within weeks as bottlers pulled or deferred orders.

Several plants responded by cutting furnace pull rates 15-25% to avoid the quality cascade that follows a chronically under-loaded forming section. On a triple-gob 8-section line running one-litre EVOO bottles at a 485 g gob weight target, a 15% pull reduction shifts gob temperature distribution faster than most operators track it. The ±2.5 g tolerance starts going wide. Base-heel wall thickness variation follows. The first call you receive is from the EVOO bottler whose automated fill-pressure sensors are flagging non-conformances, not from your own cold-end inspector. By then, a ±5 g excursion has already pushed hydraulic burst failure rates approximately three times above nominal on high-fill-pressure lines.

That's the operating reality for a Spanish container glass plant right now. Wine, olive oil, amber, Antique Green, flint: two or three furnaces, multiple campaign transitions per month, and an export profile that makes every quality excursion a commercial event rather than a rework number.

The harvest swings are not the only instability you're managing

Spain produced approximately 2.1 million tonnes of glass containers in 2023, roughly 10.5% of EU-27 output, third-largest in Europe after Germany and France, according to FEVE. Wine and olive oil bottles account for around 55% of Spanish output by weight. That product concentration is both a competitive strength and a structural liability when a single growing season cuts fill demand in half at short notice.

The energy and carbon cost structure is shifting independently of harvest cycles, and it's shifting faster than most Iberian plant P&Ls are showing. Under EU ETS Phase IV, the linear reduction factor on free allowances increased from 2.2% to 4.3% per annum effective January 2024 under Regulation EU 2023/959. A typical 300 t/day Spanish furnace running at approximately 0.42 tCO₂/tonne melt faces an uncovered EUA shortfall of 8,000-12,000 allowances per year by 2027. At the 2024 average EUA price of around €62/tonne, that is €500,000-€750,000 sitting off most conventional plant reporting structures and outside standard OEE dashboards.

The EUA shortfall is not a future risk. It is a present liability accumulating on every tonne you melt today, and most OEM-affiliated consulting engagements are not equipped to model it.

About 11 of Spain's approximately 22 operating container glass furnaces had converted to oxy-fuel by end-2024. The best-available-technology benchmark for oxy-fuel sits at 3.3-3.6 GJ/tonne of melt, against 4.2-4.8 GJ/tonne for legacy air-fuel regenerative furnaces still running at older Iberian sites. At 2024 MIBGAS reference prices, that gap is worth up to €12/tonne of melt in direct operating cost. Meanwhile, Ecovidrio, Spain's national glass collection system, recorded an 80%+ glass packaging recycling rate in 2023, enabling cullet ratios of 68-72% at Spanish furnaces. Each 10 percentage-point cullet increase reduces specific energy consumption by approximately 2.5%. But the EU Packaging and Packaging Waste Regulation, adopted in 2024, mandates separate colour-stream collection across member states by 2027, and Spanish amber cullet purity currently averages 91-93%. Running amber olive oil bottles cleanly requires purity above 95%. That gap will start showing up in your quality reports before the regulatory deadline does.

What a redox excursion on an Antique Green campaign actually costs

In 2018 I was auditing a two-furnace Iberian plant that had just commissioned an Antique Green EVOO campaign on a section that had been running flint wine bottles for two months. The forehearth arrangement made independent redox control between sections difficult, and the plant was managing it manually on shift. We caught a redox number drift to 0.12 on the AG section, below the target range of 0.18-0.28, within the first six hours of the campaign start.

Partial Cr3+ to Cr6+ conversion. Blue-green tint excursion, classified as a Type B aesthetic defect under ISO 12775. Full-pallet hold. Twenty-two hours before the section was back in colour-stable production.

Antique Green requires Cr₂O₃ dosing at 0.03-0.06% by batch weight with active redox monitoring through the early campaign hours. It also requires the hot-end superintendent to own the recipe lock, and to hold authority over batch-house chemistry adjustments on shift. No operator changes those set points without sign-off. On the plant I audited in 2018, night-shift operators were chasing colour spec off a paper sample rather than a spectrophotometer reading. The recipe lived in one man's notebook (and he was on holiday when the excursion happened, which is a separate story).

And the flint EVOO chemistry adds a parallel control problem. Premium clear olive oil bottles require Fe₂O₃ plus TiO₂ batch additions targeting less than 1% light transmission at 300 nm. A batch chemistry deviation of ±0.05% Fe₂O₃ shifts UV transmission by approximately 8%. The customer rejection on an oxidative rancidity claim arrives weeks after dispatch. By then the affected lot is in an EVOO exporter's warehouse or already in a Gulf retail channel.

The 18-22 hours you're losing to campaign transitions every month

A 10-section double-gob IS line, including older Emhart-generation machines running on retrofitted controls at several Iberian plants, requires 8-12 hours to complete a wine-to-olive-oil campaign change: blank, blow, baffle, and guide sets, plus recipe reload, lehr profile adjustment, and first-ware stabilisation. Spanish plants running mixed wine and EVOO campaigns average 4.2 changeovers per month. That is 18-22 hours of non-productive downtime per line per month. It does not appear on your IS-machine speed KPI report because the machine isn't running, not because it's running badly.

Campaign sequencing by colour family, glass weight, and mould-set compatibility can recover 4-6% OEE without capital expenditure. Sequence your Antique Green EVOO runs before your flint wine runs. Align your mould change schedule with your cullet colour stream. Lock the recipe at T-72h rather than on shift. These are scheduling decisions, not investment decisions. But they require cross-functional authority that a machine-level OEM service engagement does not hold and will not exercise.

A systemised Job Change Tool addresses the cross-shift variance directly: a versioned SKU library capturing every mould set, recipe, and forming spec, paired with a live execution checklist across the full changeover lifecycle. At most plants I audit, the variance between best and worst shift on an identical SKU runs 30-60% before anyone has formalised the process. That gap is the recoverable OEE. It's vendor-neutral in design, meaning it works across whatever OEM equipment mix your plant is running, and the gains are visible within 90 days.

If you want to map where your changeover losses actually sit, a hot end audit covers IS machine cycle analysis, forehearth profile management, and mould changeover benchmarking in a 5-10 day engagement.

The export channel is open, but so is your quality exposure

Spain's wine exports reached 2.33 billion litres valued at €3.19 billion in 2023, according to the Observatorio Español del Mercado del Vino. Bulk wine in flexitank and IBC represented 53% of volume but only 28% of value. The glass-bottled premium segment, Rioja, Cava, Priorat, carries the margin. It's also the segment most exposed to quality risk at the US channel entry point.

US TTB Regulation 27 CFR Part 4 sets a wine bottle fill tolerance of ±1.5%. A lightweight Rioja bottle at ≤300 g with shoulder wall-thickness variation exceeding ±0.3 mm risks fill-volume non-conformance at US customs inspection. That is a direct coupling between your forming precision and your customer's import clearance. Lightweight optimisation programmes that don't explicitly engineer against that specification aren't programmes. They're experiments. And the annealing lehr adds a failure mode that's easy to miss: a 360 g Burgundy wine bottle requires lehr entry at 580 ±5°C and a 10-minute soak at 560°C. A +15°C soak deviation raises residual stress above 8 nm/cm optical retardation, failing EN 1299 thermal shock. That shows up as a cold-end reject surge rather than a hot-end defect, so the corrective action gets aimed at the wrong end of the line.

The Gulf channel has a different constraint. UAE's RAK Glass produces around 250,000 t/yr domestically but is capacity-constrained on coloured glass, meaning Spain retains a premium window for Antique Green and amber EVOO flasks that regional producers can't economically match on short-run colour changes. The 2023-24 EVOO price spike to €8-9/kg ex-farm suppressed Gulf fill volumes by an estimated 18-22% year on year as buyers deferred orders or shifted to smaller pack sizes. That demand will normalise. The question is whether your plant can execute short AG and amber campaigns without the colour excursions that cost 22 hours of forming time in 2018.

For Iberian plants managing a mixed wine and olive oil asset base, asset positioning maps which SKUs, furnaces, and campaign sequences are generating your controllable losses before your next furnace campaign decision is made. If you want that analysis from a container glass consultant who has run these lines rather than sold equipment to them, that is where to start.

Look, Spain's container glass industry isn't short of capacity or technology. It is short of margin, and most of that margin is recoverable. The plants that close the campaign sequencing gap, price in their carbon liability, and formalise their changeover process in the next two years are the ones that will still be defending the premium export channels when the olive oil market stabilises.