Why This Matters

When a sugar house repeatedly struggles to process incoming syrup, the obvious conclusion is often: we need more pan capacity. That conclusion can be expensive and premature. In many sugar factories, the installed equipment is theoretically capable of handling the required throughput. The problem is that pans, strike receivers, centrifuges, steam supply and downstream equipment do not operate independently. They compete for shared resources, and disturbances in one part of the sugar house propagate into the others.

The result can look exactly like insufficient capacity: pans waiting to discharge, centrifuges overloaded at one moment and underutilised the next, steam demand peaking, strike receivers approaching their limits and operators delaying the start or progress of a batch.

The consequences continue downstream. Irregular discharge from centrifuges creates a fluctuating wet-sugar mass flow to the drying station. A continuously operated dryer must then absorb changing feed loads, making stable operating conditions and final-moisture control more difficult. Residual moisture subsequently released in storage contributes to caking and lump formation in silos. Irregular A-station operation similarly creates a fluctuating feed mass flow to the B-station, forcing the next crystallisation stage to absorb disturbances instead of operating against a steady load.

The Symptom Everyone Recognises

Sugar house scheduling optimisation exists to address that question systematically. Consider a beet sugar factory processing approximately 10,000 tonnes of beet per day. On paper, its crystallisation equipment may have enough capacity for the required sugar production. Yet during the campaign, operators repeatedly encounter periods in which the pan station becomes the apparent bottleneck.

One pan is ready to discharge, but strike receivers are too full and there is not enough capacity downstream. Another batch switches from boiling-up to pan-holding, with negative consequences on crystal content and crystal size distribution. Shortly afterwards the feed syrup tank has reached high level, forcing operators to start multiple pans at the same time. Those pans then enter their thickening phase simultaneously, causing a high steam demand peak. In parallel, centrifuges run at maximum capacity to lower strike receiver levels, creating a peak flow of wet sugar and run-off, then wait for product as the pan boiling pattern becomes uneven.

From the control room, the diagnosis seems straightforward: the sugar house needs more capacity. But equipment capacity and effective system capacity are not the same thing. What appears to be a pan, centrifuge or dryer shortage may actually be the accumulated effect of waiting, full tanks, uneven centrifuge loading, steam constraints and downstream restrictions occurring at the wrong moments. Effective capacity depends on whether the individual units are coordinated so that available pan, receiver, centrifuge, steam and downstream capacity can actually be used when needed.

What Actually Causes Pan Station Queuing

A sugar house is not simply a collection of vacuum pans. It is a network of interconnected batch and continuous processes with shared constraints. Pan station queues rarely have a single cause. They develop because several small deviations occur at the same time.

1. Upstream fluctuations force the sugar house to slow down, without a clear signal for how slow or how long.
2. Downstream restrictions, such as caking at the dryer inlet or limited conveying capacity, propagate upstream and disrupt pan scheduling.
3. Disturbances in steam supply reduce evaporation rates and extend batch duration.
4. Variations in massecuite properties change pan cycle and centrifugation times in ways that are difficult to anticipate from a fixed schedule.
5. Periodic cleaning of continuous pans temporarily reduces available crystallisation capacity at irregular intervals.
6. Centrifuges operating below normal capacity create receiver occupancy changes that cascade through subsequent pan starts.

Each event may be manageable on its own. The difficulty comes from their interaction. A delayed centrifuge changes receiver occupancy; the receiver constraint delays a pan discharge; the delayed discharge shifts the next strike start; and several shifted cycles then coincide in a high-steam-demand phase.

This becomes particularly important in sugar houses that combine continuous and batch crystallisation. A continuous pan should ideally change throughput gradually. Abrupt capacity changes can disturb massecuite quality and create equally abrupt changes in steam consumption. Batch pans meanwhile progress through discrete stages. The coordination challenge is therefore larger in mixed continuous/batch sugar houses than in a purely batch configuration.

Steam Peaks: The Cost Nobody Attributes to Scheduling

Steam demand makes the interaction especially visible. A batch vacuum pan does not consume the same amount of steam throughout its cycle. Steam demand changes as the pan moves through filling, concentration, seeding, crystal growth, tightening and discharge. If several pans enter high-demand stages simultaneously, their individual steam profiles accumulate into a large sugar-house steam peak.

The consequences extend beyond crystallisation. Evaporation and crystallisation are tightly connected through both material and energy flows. More recent sugar-industry work identifies sugar-house energy input as a key contributor to total factory energy demand, and the absence of steam flow peaks and drops is specifically identified as critical for the efficiency and technical reliability of mechanical vapour compressors.

The idea of smoothing boiling-house steam demand through scheduling is well established in the technical literature. Ramping strategies and coordinated boiling schedules reduce steam-consumption variation in pan boiling. A coordinated schedule can deliberately stagger energy-intensive phases so that instead of several pans competing for steam at once, their cycles are distributed across the available energy envelope.

How KPI-Driven Scheduling Differs From a Fixed Strike Plan

Traditional scheduling usually starts from experience. Operators know approximate pan cycle times and develop a strike sequence around them. In stable conditions, this works well. The difficulty is that a campaign rarely remains stable. A centrifuge slows down. A pan requires cleaning. Steam pressure falls. Massecuite behaviour changes. A strike takes longer than expected. The dryer becomes temporarily constrained. A fixed schedule loses relevance as soon as actual operation deviates from its assumptions.

KPI-driven scheduling works differently. Instead of asking what the original strike plan says should happen next, it asks: given the current state of the sugar house and what we predict will happen next, which action best supports our production objectives?

Model Predictive Control (MPC) is well suited to this type of problem because it repeatedly predicts future system behaviour, optimises against defined objectives and constraints, implements the next decision, and recalculates when new information becomes available. MPC and receding-horizon approaches have been applied in the literature to mixed batch/continuous sugar-end operation and scheduling with shared resources.

The Sucrosphere MPC Batch Crystallisation Scheduler applies this principle across the sugar house rather than to a single vacuum pan. Its purpose is to coordinate strike timing, steam demand and throughput while anticipating bottlenecks and synchronising batches. The KPIs reflect what matters most to the factory: maximising throughput, minimising waiting and holding time, reducing steam peaks, protecting receiver limits, maintaining centrifuge utilisation and avoiding unnecessary crystallisation interruptions.

What the Benchmark Data Shows

For a reference beet sugar factory processing approximately 10,000 t/d, the Sucrosphere Crystallisation Scheduler benchmark reports the following indicative results.

These figures are benchmark results, not universal guarantees. The achievable result depends on factory configuration, campaign length, existing automation, energy costs, product mix, equipment utilisation and, most importantly, how much scheduling inefficiency exists in the starting situation.

When You Genuinely Do Need More Capacity

Scheduling optimisation cannot create unlimited capacity. There are factories where additional equipment is the correct answer. If pans operate close to their physical capacity for sustained periods, centrifuges are continuously highly utilised without substantial waiting losses, steam is available when required, receivers and downstream systems are appropriately dimensioned, and the sugar house still cannot process the required flow, then the constraint may be structural.

Before approving a capacity investment, use plant data to answer six questions.

1. How much time do pans spend productively crystallising versus waiting?
2. How often is a pan unable to discharge because the next process step is unavailable?
3. How evenly are the centrifuges loaded across the campaign?
4. How often do multiple pans create simultaneous steam peaks?
5. How frequently do cleaning cycles or equipment disturbances propagate through the complete schedule?
6. Would a different sequence allow the same equipment to process more product?

If the analysis shows that existing assets are already being used close to their practical limits, additional capacity has a strong technical justification. But if substantial capacity is lost through waiting, conflicting strikes and poorly coordinated shared resources, adding equipment may only make a coordination problem bigger.

Frequently Asked Questions

Want to Know More?

The full technical detail, literature references and model architecture behind the Sucrosphere Crystallisation Scheduler are available in the Sucrosphere white papers. The benchmark figures cited in this article are taken from the Sucrosphere Crystallisation Scheduler whitepaper, which reports results for a 10,000 t/d beet factory reference case.

This article is part of the Sucrosphere series on sugar factory process optimisation. Related reading: How to Evaluate a Sugar Automation Vendor: The Questions That Actually Matter, and Device as a Service: How Renting Sensors Changes the Business Case for Optimisation.