A Data-First Path to Net-Zero for Manufacturing

Net-zero targets are easy to announce and hard to deliver, and most plans stall in the same place: nobody can say precisely where the energy goes today. Net zero manufacturing is a measurement problem before it is a procurement problem. Buying solar or carbon credits without an equipment-level baseline is paying to offset waste you could have eliminated for free. For Malaysian plants, the regulatory clock adds urgency: any installation consuming 21,600 GJ (about 6 GWh) or more over a rolling 12 months falls under the Energy Efficiency and Conservation Act 2024 (EECA) and must appoint a registered energy manager and report performance to the Energy Commission (Suruhanjaya Tenaga).

This is a guide to building net-zero on data you can defend, not pledges you hope to hit.

Why net zero manufacturing starts with a measured baseline

You cannot prove a reduction you never measured. Before any capital goes toward solar, storage, or offsets, establish a credible energy baseline at equipment level, not just the monthly TNB bill total. A single meter at the main intake tells you what you spent; it tells you nothing about which compressor, chiller, or furnace caused it.

A defensible baseline has three properties. It is granular (per line, per major asset, ideally per shift), it is normalised (energy per unit of output, kWh/tonne or kWh/unit, so a busy month does not masquerade as backsliding), and it is continuous (logged at minute or sub-minute resolution, not read once a quarter). This is exactly the layer CobiNeural's Insights to Energy module is built for: consumption, demand, maximum demand, power factor, and Energy Use Intensity broken down by location and by equipment.

Granular measurement starts with sub-metering the loads that dominate the bill. In most Malaysian factories, motor-driven systems (compressed air, pumps, fans) and HVAC account for the majority of electricity, and compressed air is frequently the single worst offender on a kWh-per-useful-output basis. See our guide on how sub-metering finds energy problems the main meter hides.

Reduce first, then offset: the cheapest carbon is the energy you never use

The cheapest carbon is the energy you never use, so the hierarchy is fixed: eliminate, then improve efficiency, then electrify, then procure renewables, then offset only the irreducible remainder. Plants that buy offsets before doing the first two steps overpay permanently.

Continuous monitoring is what makes the first two steps tractable. It surfaces the things a baseline alone cannot:

- Idle and off-shift load. A plant that runs 16 hours often pays for 24. Compressors cycling overnight, HVAC serving empty halls, and standby loads can be 10 to 20 percent of total consumption. Anomaly detection in CobiNeural flags the overnight floor that should not exist.
- Compressed-air leaks. A poorly maintained system can lose 20 to 30 percent of its output to leaks (U.S. Department of Energy, Improving Compressed Air System Performance). At RM0.50 to RM0.60 per kWh delivered, a single 3mm leak running continuously can cost over RM5,000 a year.
- Drifting equipment. Fouled heat exchangers, worn bearings, and degrading motors raise specific energy consumption slowly enough that nobody notices on a monthly bill. CobiNeural's Insights to Equipment module tracks motor efficiency and condition so drift is caught as a trend, not a failure.
- Maximum-demand spikes. Uncoordinated equipment startups create demand peaks that cost real money under the current TNB tariff (more below).

Offsets and on-site renewables then close the gap that efficiency cannot. Reducing first also shrinks the solar array you need, lowering capital cost before a single panel is bought.

Cut maximum demand: real money under the RP4 tariff

Maximum demand reduction delivers savings every month with little or no capital, which makes it the highest-ROI move on most net-zero roadmaps. Under TNB's RP4 tariff structure effective 1 July 2025 (Regulatory Period 4, 2025-2027), demand is billed through two separate per-kW charges applied to your monthly peak:

- General medium-voltage (C1/E1): Capacity Charge RM29.43/kW + Network Charge RM59.84/kW = RM89.27/kW per month.
- Time-of-Use medium-voltage (C2/E2): Capacity Charge RM30.19/kW + Network Charge RM66.87/kW = RM97.06/kW per month.

Source: Tenaga Nasional Berhad RP4 tariff schedule, effective 1 July 2025. The old single flat "Maximum Demand" charge has been replaced, and the bill now carries five components: Energy (sen/kWh), Capacity (RM/kW), Network (RM/kW), Retail (RM/month), and AFA (Automatic Fuel Adjustment, which replaced ICPT).

The numbers add up fast. On the General tariff, shaving 100 kW off your monthly peak saves roughly RM8,927 per month, close to RM107,000 a year for staggering startups and trimming coincident peaks, often without buying anything. On Time-of-Use, the peak window is 2:00pm to 10:00pm on weekdays; off-peak covers 10pm to 2pm on weekdays plus all weekends and public holidays. Moving discretionary load out of that window cuts both energy and demand cost.

CobiNeural tracks your live Max Demand KPI under Insights to Energy and can trigger load-shed or staggered-start sequences through the Actions module and automation layer on existing BMS, PLC, or SCADA. For the full method, see how to calculate maximum demand and our walkthrough on cutting TNB maximum demand charges.

Turn measurement into compliant reporting

The same data that drives reductions is the data auditors and customers now demand, so collect it once and report it many times. Under EECA, designated facilities must submit energy performance reports to the Energy Commission, and ISO 50001 certification requires a documented energy baseline, significant energy uses, and ongoing measurement and verification.

CobiNeural's Reporting module generates EECA-aligned outputs (anomalies, energy saving measures, and conservation reports), while the Sustainability module tracks GHG inventory across Scope 1, 2, and 3 for net-zero and ESG disclosure. The Plan and Verify module handles measurement and verification so each efficiency project's savings are quantified against the baseline rather than assumed. For how the two regimes relate, read ISO 50001 vs EECA in Malaysia and why EECA reporting starts with data.

Make energy data an operational habit, not a year-end scramble

Net-zero is sustained, not achieved once. The plants that hold their gains treat energy as a line item reviewed in the monthly operations meeting, with a named owner, alongside OEE and yield. The technology supports this: alerts route surge, anomaly, and threshold breaches to the responsible engineer over WhatsApp or email the moment they happen, so a drifting chiller is corrected in days, not discovered in next year's audit.

The sequence is the same for every site. Measure at equipment level, normalise by output, eliminate waste and demand peaks first, electrify and procure renewables for what remains, offset only the irreducible last slice, and report it all from one dataset. Skip the measurement and every later step is guesswork.

If you want to see your own plant's baseline, demand profile, and reduction opportunities laid out on real data, request a demo and we will walk through it with your numbers.