AI Logistics in Singapore: Inside a $260M Warehouse Bet

A 71,000 sq m warehouse doesn’t sound like a headline—until it’s designed to run with robotics and an automated storage and retrieval system handling about 60,000 pallet positions. That’s what CapitaLand Investment (CLI) is building in Jurong: Omega 1 Singapore, a $260 million automated logistics facility scheduled for completion in 2028.

Most companies get logistics automation wrong because they treat it like a gadget purchase: buy robots, install software, hope for efficiency. CLI’s move is a better signal to watch. It’s not “warehouse tech” for its own sake—it’s a long-term infrastructure bet built around predictable demand, labour constraints, and the reality that supply chains now compete on speed and accuracy.

This post is part of our “AI dalam Logistik dan Rantaian Bekalan” series, where we look at how AI helps with automasi gudang, ramalan permintaan, pengoptimuman laluan pengangkutan, and operational effectiveness. Omega 1 is a clean case study for what AI-driven logistics looks like when someone spends nine figures and plans for a 20-year horizon.

What the $260M Jurong facility tells us about AI logistics

The simplest read: Singapore is doubling down on high-throughput, automation-ready logistics because the old model—more headcount, more forklifts, more floor space—is hitting limits.

CLI’s fund acquired a 5.1-hectare site at 19 Gul Lane in the Jurong Industrial Estate to develop a five-storey facility. The plan includes robotics and an automated storage and retrieval system (AS/RS) designed for around 60,000 pallet positions.

Here’s the important part for business leaders: this isn’t just “more warehouse capacity.” It’s a different operating model.

Why multi-storey automation matters in Singapore

Land is scarce and expensive. Multi-storey warehouses are how Singapore scales logistics capacity without sprawling outward. But multi-storey only works well when movement is orchestrated—vertical lifts, conveyance, slotting, and replenishment have to be tightly managed.

That orchestration is where AI and automation earn their keep:

• Slotting optimisation (where inventory sits) reduces travel time and congestion.
• Predictive replenishment keeps fast movers available without overstocking.
• Workload balancing across levels and zones prevents bottlenecks.
• Exception handling (damaged goods, mispicks) becomes a measurable process, not a firefight.

If you’re running logistics in Singapore, the direction is clear: operations will increasingly be judged on throughput per square metre and accuracy per order, not only rent and headcount.

Master-lease structure: automation with guaranteed utilisation

CLI said the facility will be fully leased to Ally Logistic Property (ALP) under a master lease agreement with built-in rent escalation. From a business perspective, that’s a clue about adoption maturity.

Automation investments often fail when utilisation is uncertain—robots and AS/RS systems don’t get cheaper because your volume forecast missed. A master lease reduces volatility for the developer/investors, while ALP can focus on operations and tenant/customer acquisition.

For readers thinking about AI tools in logistics: the lesson is to pair automation with demand certainty, whether that’s long-term customer contracts, committed volumes, or strong internal consumption (e.g., retail, FMCG distribution).

The real drivers: labour, ageing demographics, and digital consumption

CLI’s leadership framed the move as a response to demand for “modern, automated logistics solutions,” driven by digitally enabled consumption, an ageing population, rising labour costs, and supply chain rationalisation.

That list isn’t PR filler. It’s a checklist for why AI in supply chain is moving from “nice to have” to “you’ll fall behind without it.”

Rising labour costs = automation becomes a planning tool, not a panic move

When labour gets tighter and more expensive, productivity improvements must be engineered into the system:

• Robotics reduce repetitive manual travel.
• Computer vision helps with scanning, dimensioning, damage detection.
• AI forecasting stabilises staffing needs by smoothing inbound/outbound peaks.

A practical stance: if your warehouse performance depends on heroic overtime during peak periods, you’re not “efficient”—you’re exposed.

Digital consumption changes the shape of demand

E-commerce and omnichannel fulfilment don’t just increase volume; they increase complexity:

• More SKUs
• Smaller order sizes
• Higher expectations on delivery time
• Greater returns volume

Automation like AS/RS pairs well with this reality because it improves pick accuracy and cycle times, especially when integrated with a strong warehouse management system (WMS).

Supply chain rationalisation is basically “less slack, more precision”

After years of disruptions, companies are redesigning networks: multi-sourcing, nearshoring where possible, and holding inventory differently. That requires better decision-making.

AI enables ramalan permintaan (demand forecasting) and inventory positioning with more granularity—by channel, by region, even by micro-seasonality.

January is a good time to revisit this because many teams are setting 2026 targets now. If your 2026 plan still treats forecasting as a monthly spreadsheet exercise, you’re budgeting for missed service levels.

What “AI in a warehouse” actually looks like (beyond robots)

People hear “automated logistics facility” and picture robots moving pallets. That’s only the visible layer. The real advantage comes from connected decision systems.

Here’s a useful mental model: modern smart logistics hubs run on four layers.

1) Sense: collect clean operational data

You can’t optimise what you can’t measure. High-performing sites typically capture:

• Real-time inventory accuracy
• Location-level movement history
• Equipment utilisation (lifts, conveyors, shuttles)
• Order cycle times by wave/batch
• Exceptions (short picks, damages, delays)

2) Decide: AI models turn data into actions

This includes:

• Demand forecasting (SKU-level, channel-level)
• Dynamic slotting (place items where they reduce travel)
• Wave planning (group orders to reduce congestion)
• Predictive maintenance (service equipment before it fails)

One-liner worth keeping: Automation moves goods; AI decides how work should flow.

3) Act: automation executes consistently

AS/RS, robotics, and conveyor systems execute decisions with less variability than manual operations. Consistency is underrated—consistent performance makes service levels predictable.

4) Learn: continuous improvement becomes systematic

When the system captures outcomes (late orders, missed picks, queue build-up), teams can improve rules, retrain models, and redesign layouts based on evidence.

If you want a KPI that reflects this maturity, track variance: variance in cycle time, variance in labour hours per 1,000 lines, variance in dock-to-stock time.

How Singapore businesses can apply these lessons without a $260M budget

Not everyone is building a five-storey automated warehouse. But the underlying playbook is available to mid-sized distributors, manufacturers, and 3PLs—if you scope it correctly.

Start with “where is the bottleneck?” not “what tech should we buy?”

In many Singapore operations I’ve seen, the bottleneck is one of these:

• Inbound receiving and putaway delays
• Poor slotting (fast movers too far, slow movers too close)
• Picking congestion during peaks
• Inventory accuracy issues causing rework
• Manual planning (waves, staffing, replenishment)

Pick one bottleneck and solve it end-to-end.

A practical 90-day roadmap for AI dalam logistik

If you want something you can execute this quarter:

1. Weeks 1–2: Baseline your process
   • Measure order cycle time, pick accuracy, dock-to-stock, labour hours per order.
2. Weeks 3–6: Fix data quality and visibility
   • Clean SKU masters, standardise location naming, enforce scanning discipline.
3. Weeks 7–10: Deploy one optimisation use case
   • Example: slotting optimisation or demand forecasting for top 200 SKUs.
4. Weeks 11–13: Operationalise
   • Put the output into daily routines (re-slot weekly, forecast weekly, review exceptions).

This isn’t about fancy dashboards. It’s about repeatable decisions.

What to ask vendors or internal teams (so you don’t waste money)

When evaluating AI tools for supply chain and warehouse automation, ask:

• What decisions will the model make, and how often? (hourly, daily, weekly)
• What data is required, and what’s optional?
• How do we handle exceptions? (damages, returns, urgent orders)
• What does success look like in numbers? (e.g., -15% travel time, +0.3% accuracy)
• How will this integrate with our WMS/ERP?

If the answers are vague, expect vague results.

People also ask: common questions about automated logistics hubs

Is AS/RS only for very large warehouses?

No. AS/RS makes the most sense when you have high SKU counts, tight space, and frequent picking/replenishment. Smaller sites can start with mini-load systems or targeted automation in high-velocity zones.

Does warehouse automation reduce headcount?

Sometimes, but the bigger effect is reallocating labour from travel and manual handling to exception resolution, quality checks, and coordination. The best sites don’t just “cut people”—they reduce chaos.

What’s the ROI benchmark for automation?

ROI depends on volume stability, labour costs, space constraints, and service-level penalties. A useful rule: if your peaks drive overtime and missed SLAs, automation and AI planning often pay back faster than expected.

What Omega 1 Singapore signals for 2026 planning

CLI has deployed about $500 million into logistics developments across South-east Asia in the past two years, and this Jurong acquisition shifts its fund portfolio to 55% in Singapore. That’s a strong vote for Singapore as an automation-first logistics base.

For operators and business owners, the implication is straightforward: customers will get used to faster, more accurate fulfilment—and they’ll expect it from you too. Competing on “we try our best” won’t survive against supply chains engineered for precision.

If you’re following our AI dalam Logistik dan Rantaian Bekalan series, treat this as your prompt to audit your own operations:

• Where does work pile up every week?
• Which decisions are still made by gut feel?
• Which part of your warehouse would improve most from automation + better planning?

The next wave of advantage won’t come from building the biggest warehouse. It’ll come from running the smartest one.