Problem Statements
Explore the problem statements from Sakthi's diverse industrial sectors and choose the one that inspires you. Build innovative solutions for a better tomorrow.
Problem Statements (4 Found)
Automatic Alloy Addition Calculation from Spectrometer Composition
During melting, spectrometer results indicate the actual chemical composition of molten metal. Determining corrective alloy additions requires calculations based on target chemistry, melt weight and recovery factors. Manual calculation can introduce delay and variation. The challenge is to develop a system that automatically recommends alloy additions from spectrometer results.
AI-Based Induction Furnace Power Optimization System
Induction furnace parameters such as power, current, voltage, temperature, charge weight and melting time are monitored separately, making it difficult to determine the optimum power profile for each heat. The challenge is to use real-time and historical heat data to recommend power adjustments, predict melting time, target temperature and energy consumption, and improve process efficiency.
Development of a Data-Driven Dynamic Soybean Procurement Strategy for ABT-CPD
Soybean is the primary raw material for ABT-CPD, and its market price is highly variable. Soybean is available through both domestic and international markets, with significant variations in price, seasonality, availability and procurement economics. Historical domestic and international soybean price data, along with ABT-CPD’s procurement data, provide an opportunity to develop a scientific and data-driven procurement strategy. The challenge is to develop a decision-support system that can analyze historical and current market trends and recommend when to buy, how much to buy, and whether domestic or international sourcing is economically preferable, while considering inventory, lead time, landed cost and business requirements. The objective is to reduce procurement cost and price risk while ensuring uninterrupted availability of quality soybean.
ABT-CPD Zero Customer Complaint – AI-Based Complaint Analytics, Root Cause Identification & Prediction System
ABT-CPD has accumulated approximately 19 years of customer complaint records, covering different products, customers and types of quality and service-related complaints. Complaints include product quality deviations, physical defects, packaging issues, contamination/foreign matter, infestation, sensory characteristics, weight/quantity issues, process-related defects and other customer-specific concerns. At present, the historical data represents a valuable but largely untapped source of information. The challenge is to analyse this historical database using data analytics and AI to identify recurring patterns, major root causes, high-risk products/processes, ineffective or recurring corrective actions and early warning indicators. The ultimate objective is to develop a system capable of identifying the probability and likely nature of future customer complaints and recommending preventive actions before the product reaches the customer.