Plant productivity is not a target—it is a consequence of visibility into three mechanics: what your bottleneck machine actually produces (Overall Equipment Effectiveness), what each hour of unplanned downtime costs your cash flow, and how much energy you burn per unit of output.
Most Indian SME manufacturers measure none of these with discipline. They know production volume and despatch dates, but not OEE, not the true cost of a breakdown, and not whether a 15% energy spend increase in the last quarter was driven by more output or lower efficiency. This playbook walks you through diagnosis on one machine, how to cost downtime honestly, and the first 10 recovery steps that typically yield 8–15 percentage-point OEE gains within 90 days.
Advisory
Overall Equipment Effectiveness = (Availability % × Performance % × Quality %) ÷ 100. Availability = actual run time ÷ scheduled time. Performance = actual output ÷ theoretical maximum per minute. Quality = good units ÷ total units produced. Example: a bottleneck injection-mould running 7 am–5 pm (8 hours scheduled) actually runs 6.5 hours (breakdowns + changeover = 1.5 hours). Availability = 81.25%. It produces 120 units/hour but the standard is 150/hour (congestion, operator pace). Performance = 80%. Of 780 units, 42 are defective (5.4% scrap). Quality = 94.6%. OEE = 0.8125 × 0.80 × 0.946 = 61.4%. Indian SME benchmark is 55–65%; world-class is 85+%. You must log this weekly on a single sheet (date, scheduled hours, actual run hours, output, defects) to see drift and isolate the worst month.
Unplanned downtime costs more than lost production. Example: a ₹1.5 crore/year plastic-injection shop running at ₹25 lakh per day revenue. One unexpected breakdown = 8 hours lost. Revenue loss = ₹25,00,000 ÷ 24 × 8 = ₹8,33,000. But despatch delays may trigger customer penalties (₹50,000–2,00,000 per day late), rework of inventory held, and overtime labour (₹30,000–60,000 for 8 hours). Real cost per 8-hour breakdown = ₹9–12 lakhs. A preventive maintenance programme (lubrication, wear-part replacement, calibration) costing ₹4–6 lakhs per quarter and cutting breakdowns from 2–3 per month to 1 every 6 months justifies itself in 6 weeks. Log every unplanned stop: date, duration, cause (bearing wear, tooling fault, electrical), and estimated forgone revenue. This disciplines decisions on whether to hire a maintenance technician (₹35,000–45,000/month all-in) or buy a ₹15 lakh CNC monitoring system.
Energy cost per unit = total monthly electricity spend ÷ total units produced. If your bill (all charges, not just kWh) is ₹4,00,000/month and output is 40,000 units, cost = ₹10 per unit. Now break the bill: base consumption (₹2,50,000), demand surcharge for peak load (₹80,000), and
Frequently asked questions
OEE = (Availability % × Performance % × Quality %) ÷ 100. Availability is actual run time ÷ scheduled time, Performance is actual output ÷ theoretical maximum, and Quality is good units ÷ total units produced. Indian SME benchmark is 55–65%; world-class is 85+%.
Log weekly data on a single sheet: date, scheduled hours, actual run hours, total output, and defects. Calculate the three percentages, multiply them together, and divide by 100. Track monthly drift to isolate problem areas.
The 10-point playbook includes costing downtime per hour honestly, diagnosing availability/performance/quality gaps, and executing shop-floor interventions tied to real data. Typical gains are 8–15 percentage points within 90 days.