Salem Steel: How India's Secondary Steel Capital is Using AI to Compete with Integrated Mills

Salem district has the highest concentration of steel rolling mills outside of Raipur and Surat. The city's 200+ mills process 2 lakh tonnes of steel monthly — predominantly TMT bars, MS angles, and flats — supplied to the construction industry across Tamil Nadu, Kerala, and Karnataka.

It is a competitive industry with thin margins and relentless cost pressure. The large integrated steel producers (JSW, Tata, SAIL) have cost advantages in raw material and energy. Salem's secondary mills compete on flexibility, relationships, and — increasingly — quality.

The Secondary Steel AI Advantage

Secondary steel rolling mills have a structural challenge: they process scrap and billet inputs with variable chemistry and quality. This input variability — which integrated mills avoid by controlling their own upstream processes — creates quality challenges that AI is uniquely suited to address.

Input Material AI Analysis. Before scrap is charged to the furnace, AI optical spectrometry can estimate its chemical composition — informing the melting and alloying process to achieve target steel chemistry with less trial and error. Salem mills using this approach report 15-20% reduction in alloying cost and more consistent output chemistry.

Rolling Process AI. The rolling process turns heated billet into finished bar through a sequence of rolling stands. AI-controlled speed and gap settings at each stand — responding to real-time temperature and dimensional measurements — produce more consistent finished dimensions and fewer out-of-tolerance bars. Scrap rates drop from 6-8% (manual control) to 2-3% (AI control).

Surface Inspection AI. TMT bar surface quality — free from seams, laps, and surface cracks — is a structural requirement. AI vision systems inspect every bar at full production speed, catching defects that manual inspection misses under the high-temperature, high-speed production conditions. One Salem mill reduced their customer quality complaints by 84% in 6 months after deploying inspection AI.

Energy: The Biggest Cost Lever

Energy typically represents 25-35% of operating costs for a Salem rolling mill. The reheating furnace — which brings cold billet to rolling temperature (1,100-1,200°C) — is the largest energy consumer. Manual furnace operation relies on operator experience and periodic temperature measurements. AI combustion control monitors furnace temperature continuously across multiple zones, optimizing fuel injection and air ratios in real time.

The result: fuel consumption reduction of 18-24% with no reduction in throughput. For a mill with ₹3 crore monthly energy bills, this saves ₹54-72 lakh per year — easily justifying the technology investment within 6-12 months.

Predictive Maintenance: Avoiding the ₹20 Lakh Shutdown

An unplanned rolling mill shutdown costs ₹15-25 lakh per day in lost production and restart costs. Equipment failures — particularly of rolling stands, water cooling systems, and reheating furnace components — are the primary cause. AI predictive maintenance monitors vibration patterns, bearing temperatures, and motor current signatures to detect emerging failures 2-4 weeks before they cause shutdowns.

Salem mills deploying predictive AI report 60-70% reduction in unplanned downtime — translating to 6-10 additional production days per year and a proportional revenue increase.

The Future: BIS Certification and Premium Market Access

Salem mills that achieve consistent AI-enabled quality documentation are better positioned for BIS certification — which opens government infrastructure contracts and export markets. This is the long-term strategic value of AI quality systems: not just efficiency today, but market access tomorrow.