Time Series Isn’t Enough: How Graph Neural Networks Change Demand Forecasting

in supply-chain planning has historically been handled as a time-series downside.

• Every SKU is modeled independently.
• A rolling time window (say, final 14 days) is used to foretell tomorrow’s gross sales.
• Seasonality is captured, promotions are added, and forecasts are reconciled downstream.

And but, regardless of more and more subtle fashions, the standard issues persist:

• Continual over-and under-stocking
• Emergency manufacturing adjustments
• Extra stock sitting within the flawed place
• Excessive forecast accuracy on paper, however poor planning outcomes in apply

The problem is that demand in a provide chain will not be impartial. It’s networked. For example, that is what simply 12 SKUs from a typical provide chain appear like if you map their shared vegetation, product teams, subgroups, and storage places.

So when demand shifts in a single nook of the community, the consequences are felt all through the community.

On this article, we step exterior the model-first pondering and have a look at the issue the way in which a provide chain really behaves — as a linked operational system. Utilizing an actual FMCG dataset, we present why even a easy graph-based neural community(GNN) essentially outperforms conventional approaches, and what which means for each enterprise leaders and information scientists.

An actual provide chain experiment

We examined this concept on an actual FMCG dataset (SupplyGraph) that mixes two views of the enterprise:

Static supply-chain relationships

The dataset has 40 energetic SKUs, 9 vegetation, 21 product teams, 36 sub-groups and 13 storage places. On common, every SKU has ~41 edge connections, implying a densely linked graph the place most SKUs are linked to many others by shared vegetation or product teams..

From a planning standpoint, this community encodes institutional data that always lives solely in planners’ heads:

“If this SKU spikes, these others will really feel it.”

Temporal operational indicators and gross sales outcomes

The dataset has temporal information for 221 days. For every SKU and every day, the dataset consists of:

• Gross sales orders (the demand sign)
• Deliveries to distributors
• Manufacturing facility items points
• Manufacturing volumes

Right here is an outline of the 4 temporal indicators driving the provision chain mannequin:

As could be noticed, virtually half of the SKU-Day combos have zero gross sales. The implication being a small fraction of SKUs drives many of the quantity. It is a traditional “Intermittent Demand” downside.

Additionally, manufacturing happens in rare, massive batches (lumpy manufacturing). Downstream supply is far smoother and extra frequent (low sparsity) implying the provision chain makes use of vital stock buffers.

To stabilize GNN studying and deal with excessive skew, all values are remodeled utilizing log1p, an ordinary apply in intermittent demand forecasting.

Key Enterprise Metrics

What does an excellent demand forecast appear like ? We consider the mannequin based mostly on two metrics; WAPE and Bias

WAPE — Weighted Absolute Share Error

WAPE measures how a lot of your whole demand quantity is being mis-allocated. As an alternative of asking “How flawed is the forecast on common throughout all SKUs?“, WAPE asks the query supply-chain planners really care about within the situation of intermittent demand: “Of all SKU models that had been moved by the provision chain to fulfill demand, what fraction was mis-forecast?“

This issues as a result of errors on high-volume SKUs value way over errors on long-tail gadgets. A ten% miss on a high vendor is costlier than a 50% miss on a gradual mover. So WAPE weights the SKU-days by quantity offered, and aligns extra naturally with income affect, stock publicity, plant and logistics utilization (and could be additional weighted by value/SKU if required).

That’s why WAPE is broadly most well-liked over MAPE for intermittent, high-skew demand.

[
text{WAPE} =
frac{sum_{s=1}^{S}sum_{t=1}^{T} left| text{Actual}_{s,t} – text{Forecast}_{s,t} right|}
{sum_{s=1}^{S}sum_{t=1}^{T} text{Actual}_{s,t}}
]

WAPE could be calculated at totally different ranges — product group, area or whole enterprise — and over totally different durations, corresponding to weekly or month-to-month.

It is very important notice that right here, WAPE is computed on the hardest potential stage — per-SKU, per-day, on intermittent demand — not after aggregating volumes throughout merchandise or time. In FMCG planning apply, micro-level SKU-daily WAPE of 60–70% is commonly thought of acceptable for intermittent demand, whereas <60% is taken into account production-grade forecasting.

Forecast Bias — Directional Error

Bias measures whether or not your forecasts systematically push stock up or down. Whereas WAPE tells you how flawed the forecast is, Bias tells you how operationally costly it’s. It solutions a easy however important query: “Will we persistently over-forecast or under-forecast?”. As we are going to see within the subsequent part, it’s potential to have zero bias whereas being flawed more often than not. In apply, constructive bias ends in extra stock, greater holding prices and write-offs whereas detrimental bias results in stock-outs, misplaced gross sales and repair penalties. In apply, a bit of constructive bias (2-5%) is taken into account production-safe.

[ text{Bias} = frac{1}{S} sum_{s=1}^{S} (text{Forecast}_s – text{Actual}_s) ]

Collectively, WAPE and Bias decide whether or not a mannequin is not only correct, its forecasts are operationally and financially usable.

The Baseline: Forecasting With out Construction

To determine a floor flooring, we begin with a naïve baseline, which is “tomorrow’s gross sales equal at present’s gross sales”.

[ hat{y}_{t+1} = y_t ]

This method has:

• Zero bias
• No community consciousness
• No understanding of operational context

Regardless of its simplicity, it’s a robust benchmark, particularly over the brief time period. If a mannequin can not beat this baseline, it’s not studying something significant.

In our experiments, the naïve method produces a WAPE of 0.86, that means almost 86% of whole quantity is misallocated.

The bias of zero will not be an excellent indicator on this case, since errors cancel out statistically whereas creating chaos operationally.

This results in:

• Firefighting
• Emergency manufacturing adjustments
• Expediting prices

This aligns with what many practitioners expertise: Easy forecasts are secure — however flawed the place it issues.

Including the Community: Spatio-Temporal GraphSAGE

We use GraphSAGE, a graph neural community that enables every SKU to combination data from its neighbors.

Key traits:

• All relationships are handled uniformly.
• Info is shared throughout linked SKUs.
• Temporal dynamics are captured utilizing a time sequence encoder.

This mannequin doesn’t but distinguish between vegetation, product teams, or storage places. It merely solutions the important thing query:

“What occurs when SKUs cease forecasting in isolation?”

Implementation

Whereas I’ll dive deeper into the information science behind the function engineering, coaching, and analysis of GraphSAGE in a subsequent article, listed below are a few of the key rules to grasp:

• The graph with its nodes and edges types the static spatial options.
• The spatial encoder part of GraphSAGE, with its convolutional layers, generates spatial embeddings of the graph.
• The temporal encoder (LSTM) processes the sequence of spatial embeddings, capturing the evolution of the graph during the last 14 days (utilizing a sliding window method).
• Lastly, a regressor predicts the log1p-transformed gross sales for the subsequent day.

An intuitive analogy

Think about you’re attempting to foretell the worth of your home subsequent month. The worth isn’t simply influenced by the historical past of your individual home — like its age, upkeep, or possession data. It’s additionally influenced by what’s taking place in your neighborhood.

For instance:

• The situation and costs of homes much like yours (related development high quality),
• How well-maintained different homes in your space are,
• The provision and high quality of shared providers like faculties, parks, or native regulation enforcement.

On this analogy:

• Your home’s historical past is just like the temporal options of a specific SKU (e.g., gross sales, manufacturing, supply historical past).
• Your neighborhood represents the graph construction (the sides connecting SKUs with shared attributes, like vegetation, product teams, and so forth.).
• The historical past of close by homes is just like the neighboring SKUs’ options — it’s how the habits of different related homes/SKUs influences yours.

The aim of coaching the GraphSAGE mannequin is for it to be taught the operate f that may be utilized to every SKU based mostly on its personal historic options (like gross sales, manufacturing, manufacturing facility points, and so forth.) and the historic habits of its linked SKUs, as decided by the sting relationships (e.g., shared plant, product group, and so forth.). To depict it extra exactly:

embedding_i(t) =
  f( own_features_i(t),
     neighbors’ options(t),
     relationships )

the place these options come from the SKU’s personal operational historical past and the historical past of its linked neighbors.

The End result: A Structural Step-Change

The affect is sort of exceptional:

In sensible phrases:

• The naïve method misallocates almost 86% of whole demand quantity
• GraphSAGE reduces this error by ~27%

The next chart reveals the precise vs predicted gross sales on the log1p scale. The diagonal pink line depicts excellent forecast, the place predicted = precise. As could be seen, many of the excessive quantity SKUs are clustered across the diagonal which depicts good accuracy.

From a enterprise perspective, this interprets into:

• Fewer emergency manufacturing adjustments
• Higher plant-level stability
• Much less handbook firefighting
• Extra predictable stock positioning

Importantly, this enchancment comes with none extra enterprise guidelines — solely by permitting data to circulate throughout the community.

And the bias comparability is as follows:

At beneath 5%, the gentle forecasting bias GraphSAGE introduces is properly inside production-grade limits. The next chart depicts the error within the predictions.

It may be noticed that:

• Error is negligible for many of the forecasts. Recall from the temporal evaluation that sparsity in gross sales is 46%. This reveals that the mannequin has discovered this, and is accurately predicting “Zero” (or very near it) for these SKU-days, creating the height on the middle.
• The form of the bell curve is tall and slim, which signifies excessive precision. Most errors are tiny and clustered round zero.
• There’s little skew of the bell curve from the middle line, confirming the low bias of 4.5% we calculated.

In apply, many organizations already bias forecasts intentionally to guard service ranges, somewhat than danger stock-outs.

Let’s have a look at the affect on the SKU stage. The next chart reveals the forecasts for the highest 4 SKUs by quantity, denoted by pink dotted strains, in opposition to the actuals.

Just a few observations:

• The forecast is reactive in nature. As marked in inexperienced circles within the first chart, the forecast follows the precise on the way in which up, and in addition down with out anticipating the subsequent peak properly. It is because GraphSAGE considers all relations to be homogeneous (equally necessary), which isn’t true in actuality.
• The mannequin under-predicts excessive spikes and compresses the higher tail aggressively. GraphSAGE prefers stability and smoothing.

Here’s a chart exhibiting the efficiency throughout SKUs with non-zero volumes. Two threshold strains are marked at WAPE of 60% and 75%. 3 of the 4 highest quantity SKUs have a WAPE < 60% with the fourth one simply above. From a planning perspective, it is a strong and balanced forecast.

Takeaway

Graph neural networks do greater than enhance forecasts — they modify how demand is known. Whereas not excellent, GraphSAGE demonstrates that construction issues greater than mannequin complexity.

As an alternative of treating every SKU as an impartial downside, it permits planners to purpose over the provision chain as a linked system.

In manufacturing, that shift — from remoted accuracy to network-aware decision-making — is the place forecasting begins to create actual financial worth.

What’s subsequent? From Connections to That means

GraphSAGE confirmed us one thing highly effective: SKUs don’t stay in isolation — they stay in networks.
However in our present mannequin, each relationship is handled as equal.

In actuality, that isn’t how provide chains work.

A shared plant creates very totally different dynamics than a shared product group. A shared warehouse issues in a different way from a shared model household. Some relationships propagate demand shocks. Others dampen them.

GraphSAGE can see that SKUs are linked — but it surely can not be taught how or why they’re linked.

That’s the place Heterogeneous Graph Transformers (HGT) are available in.

HGT permits the mannequin to be taught totally different behaviors for various kinds of relationships — letting it weigh, for instance, whether or not plant capability, product substitution, or logistics constraints ought to matter extra for a given forecast.

Within the subsequent article, I’ll present how shifting from “all edges are equal” to relationship-aware studying unlocks the subsequent stage of forecasting accuracy — and improves the standard of forecast by including that means to the community.

That’s the place graph-based demand forecasting turns into actually operational.

Join with me and share your feedback at www.linkedin.com/in/partha-sarkar-lets-talk-AI

Reference

SupplyGraph: A Benchmark Dataset for Supply Chain Planning using Graph Neural Networks : Authors: Azmine Toushik Wasi, MD Shafikul Islam, Adipto Raihan Akib

_{Photos used on this article are synthetically generated. Charts and underlying code created by me.}