Beyond visibility: Turning supply chain data into faster, better decisions
More than half of Australian businesses changed how they operate in the first half of 2026 as fuel and freight costs soared. With pressure on supply chains continuing, Supply Chain Insights explores how leading organisations are turning data into faster, more confident decisions, and using this to strengthen operations against evolving disruption.
The bottleneck
Andre Tibyrica, Supply Chain Director at TMX Transform, thinks the gap between data-rich businesses and decisive ones comes down to time.
“The gap is the ability to shorten the time between insight and action. The temptation can be to harness more data, more dashboards, reporting and analytics, though the speed at which you can analyse, test options and make decisions is more powerful,” he says.
That speed is vital for structural decisions such as warehouse leases, 3PL agreements and carrier contracts, which often run on three-to-five-year cycles. Speed matters because your supply chain is how you deliver on customer experience. If you can’t make decisions quickly, with a clear view of the trade-offs, you can miss windows of opportunity,” Andre explains.
In his experience, the slowdown often happens after the analysis is done. “Someone might analyse a change and find it will halve inventory. But then stakeholders want to know what it does to cost, capacity and service. Without a system that answers all those questions at once, decisions get stuck in internal governance.”
Bringing risk into the picture can speed things up. “The human brain thinks in averages,” Andre notes. “Showing the inventory saving alongside the risk to service levels – for example, a 30% chance of missing them – creates a whole different feeling for the person approving that decision.”
Michael Bradshaw, Vice President, Solution Design and Sales – APAC, MEA at Dematic, agrees the constraint has moved.
“Accessing relevant data and performing detailed modelling took significant time in the past. AI tools have changed that,” Michael says. “As a result, speed of response is increasingly constrained by the time needed to evaluate considerations such as customer commitments, commercial trade-offs, organisational priorities and risk management.”
The final call, he continues, still rests with people. “Technology can provide supporting information quickly, but experienced leadership is still often required to determine which direction to take.”
Where AI is delivering today
AI sits at the centre of most conversations about faster decision-making, and all three experts see real value in the technology when it’s pointed at the right problem.
For Michael, AI’s strength is its ability to look ahead. “AI is opening up new possibilities in turning large amounts of data into information and insights that support better and faster decisions. It can also be used to predict what might happen, as well as reporting what happened in the past,” he says.
“For planning and forecasting, AI can easily automate many routine workflows and decision points. However, many business decisions involve dealing with incomplete information, making commercial trade-offs, respecting organisational politics and understanding appetite for corporate risk. AI will be an excellent advisor and decision-support tool, while experienced supply chain and business leaders remain ultimately accountable for strategic decisions.”
According to Manhattan’s Raghav Sibal, AI is already proving its worth where the problem is well defined. “We’re seeing that in demand forecasting, inventory optimisation, replenishment and identifying potential disruptions or constraints before they become bigger issues,” he says.
He also urges some caution. “We need to be careful about being too quick to consider AI the answer to every supply chain problem. Adding another AI tool doesn’t necessarily result in a better decision. You still need good data, connected systems and a clear understanding of what you are trying to improve,” Raghav explains.
“The real opportunity is using AI to help people work through complexity faster, whether that’s identifying an exception, recommending an action or helping a planner understand the implications of a decision.”
TMX Transform’s Andre sees AI doing a good job in commodity forecasting, where it can interpret market signals from news and weather, and in inventory optimisation, where “most tools using traditional algorithms take very big shortcuts.” He notes it has further to go in decisions that cut across several dimensions at once.
“When a decision involves inventory, cost, service and capacity, AI handles parts of it, but tends to see things in black and white. People are better at working in the grey, where several factors are at play.”
Test before you commit
Modelling gives businesses a way to see the likely impact of a decision before they commit capital, people or customer promises to it.
“A great example is the simulation of distribution centres and the automation deployed inside them,” Dematic’s Michael says. “Before investing millions of dollars in automated systems, organisations can create a digital model of a proposed operation and test it using representative inventory and order data.
“Using simulation, we can validate throughput, capacity, labour requirements and operating costs, while also performing sensitivity testing across a range of assumptions. The value of this approach extends beyond validation. It enables optimisation, reduces implementation risk and increases confidence in major investment decisions before they are made.”
With capital investment under pressure across Australian businesses this year, the ability to test a major investment before signing off on it is increasingly valuable.
Raghav says the same thinking applies to day-to-day operations. “Businesses can use their own operational data to model changes before introducing them into a live environment.”
When considering what a retailer facing a busy peak season might need to know – what would happen if order volumes increased during peak, if inventory was positioned differently, or if it changed the way orders were prioritised and fulfilled – Raghav points to modelling. “Modelling allows the business to test different options and understand the likely impact on capacity, throughput, service levels and cost before making the change.”
He notes the real value lies in seeing the flow-on effects. “Supply chains are interconnected, so a decision that improves one metric can create pressure somewhere else. Testing those trade-offs using real operational data gives businesses much greater confidence before they make a change.”
At network level, Andre describes working with a fast-growing homewares retailer that was close to running out of capacity, where every shift in the forecast affected its capital programme for automation.
“We created a digital twin of their network. Whenever circumstances change, a data workflow runs quickly through the system and covers all the dimensions they need to make a decision – service, cost, inventory, capacity. It’s almost at the press of a button,” he explains. “It has become an ongoing part of budgeting, capacity planning and the capital plan, in a much more agile way than we see most companies doing.”
From visibility to response
With supplier delivery times still stretching, knowing about a problem is only the first step.
“Data and visibility are important, but they only get you so far. Knowing that something has gone wrong is one thing. Understanding the impact and the best way to respond is another,” Raghav says.
“If a shipment is delayed, you need to understand which orders are affected, where else inventory is available, whether those orders can be fulfilled differently and what that means for transport and customer commitments. The faster you can move from seeing the problem to understanding the impact and deciding what to do next, the more opportunity you have to minimise disruption.”
Michael says data from the physical operation is now closing the loop between planning and execution.
“Historically, planning systems and operational systems were often disconnected. Planning generated forecasts and targets, but actual operating outcomes could be quite different, and results were not always captured or fed back quickly enough to influence decisions.”
He notes the gap is shrinking as warehouse and automation systems capture more of what happens on the floor. “Today, modern WMS, WCS, SCADA and automated systems generate huge amounts of operational data. Actual throughput, equipment performance and labour productivity data available in real time can be used to continuously adjust workload balancing, labour plans and inventory positioning.”
Michael sees this going further. “The long-term vision is for real-world performance data to feed directly into planning models and digital twins, enabling organisations to make faster and better-informed decisions based on what is actually happening.”
Andre thinks a fast response starts well before anything goes wrong. “You need to know what you’re protecting against and what your trigger points are, so a red flag pops up proactively,” he explains. “Then you need repeatable processes to turn the data into a decision, so by the time you identify the issue, you already know the direction to take.”
People and machines
As technology takes on more of the analysis, the role of people in supply chain decisions is shifting.
Andre has seen that MRP and forecasting are already well served by automation, but says people still add context the system can’t see. “People know the business is going
to run this promotion, they know a specific shipping lane is very congested, they know the weather at a certain time of year has a big impact,” he says.
Bigger calls, such as where a DC network should sit, benefit from a hybrid approach. “The goal should be to remove humans from decisions machines can reliably make, so people can spend more time on the exceptions, trade-offs and strategic decisions where human context creates value,” Andre explains.
Raghav sees AI taking on the legwork. “A planner can spend a lot of time gathering information, comparing options and working out where they need to focus. AI and decision intelligence can do much of that work much faster, surfacing exceptions and recommending possible actions,” he says.
“The technology might tell you what it recommends and why, but there will still be situations where a planner knows something the system does not. For me, that is where the value is – giving people better information and more time to apply it.”
Michael agrees the balance is moving, “although perhaps not in the way many people initially expected.”
“Historically, a significant portion of knowledge work involved gathering information, analysing data and producing reports. Today, AI can automate much of that effort, allowing people to focus on higher-value activities that better leverage their expertise and experience,” he explains.
“Organisations will continue to rely on people for decisions that require strategic thinking, creativity, judgement, leadership and accountability. The future is about supporting and supplementing people with AI so they can make better decisions, faster.”
This article was originally published in Supply Chain Insights on 29 October 2026.