Change Management In Agile Environments

SPC – The Language of Process Stability We often hear: “The machine is running fine.” But how do we prove it? That’s where SPC (Statistical Process Control) steps in. SPC uses data + statistics to check whether a process is: Stable (predictable, under control) Capable (meeting customer requirements) Step 1: Define the Specification Suppose we are making a motorcycle frame tube joint. Required specification = Diameter = 50.00 mm ± 0.20 mm That means: LSL (Lower Spec Limit) = 49.80 mm USL (Upper Spec Limit) = 50.20 mm --- Step 2: Collect Data from Production We take 5 samples every shift. Example readings: 50.05, 50.02, 49.98, 50.07, 50.01 --- Step 3: Calculate the Average (X̄) and Range (R) Average (X̄) = (50.05 + 50.02 + 49.98 + 50.07 + 50.01) ÷ 5 = 250.13 ÷ 5 = 50.026 mm Range (R) = Highest – Lowest = 50.07 – 49.98 = 0.09 mm This tells us the process is not fluctuating wildly. --- Step 4: Estimate Variation (σ) For simplicity, assume σ = R ÷ d2 (where d2 is a statistical factor). For sample size 5, d2 = 2.326. σ = 0.09 ÷ 2.326 ≈ 0.039 mm --- Step 5: Check Process Capability (Cp & Cpk) 1. Cp (Potential Capability): Formula = (USL – LSL) ÷ (6σ) = (50.20 – 49.80) ÷ (6 × 0.039) = 0.40 ÷ 0.234 = 1.71 Means the process has the potential to meet specs. --- 2. Cpk (Actual Capability): Formula = min[(X̄ – LSL) ÷ (3σ), (USL – X̄) ÷ (3σ)] = min[(50.026 – 49.80) ÷ (0.117), (50.20 – 50.026) ÷ (0.117)] = min[(0.226 ÷ 0.117), (0.174 ÷ 0.117)] = min[1.93, 1.49] = 1.49 Means the process is well within limits, slightly shifted but safe. --- Step 6: Interpret with a Table Cp / Cpk Value Meaning < 1.00 Not capable – high risk of defects 1.00 – 1.33 Marginal – needs improvement 1.33 – 1.67 Capable – industry acceptable > 1.67 World class, highly capable --- Final Takeaway Cp = What the process could achieve Cpk = What the process is actually delivering For our motorcycle frame: Cp = 1.71 → Machine/process is excellent. Cpk = 1.49 → Process is stable, safe, and customer won’t see defects. SPC is not just math – it’s an early warning system to avoid costly rework or recalls. In short: SPC = Early warning system before customers complain. #Quality #SPC #Manufacturing #LeanSixSigma #VIPtalks

Part 6: Practical Implementation : Purge, Testing, and Lifecycle Management ICH M7 becomes meaningful only when risk assessment is translated into process control, purge logic, and targeted testing. A mutagenic impurity does not always need routine final API testing if its formation, fate, and removal are scientifically understood and carryover above the acceptable limit is negligible. Practical control examples A. Nitrosamine control Risk drivers: secondary/tertiary amines, nitrite source, acidic or enabling conditions. Controls: eliminate nitrite where possible, control nitrite in inputs and water, avoid vulnerable amines in nitrosating environments, control pH and hold time, and use confirmatory LC-MS/MS or GC-MS where needed. B. Alkyl sulfonate control Risk drivers: sulfonic acid + methanol/ethanol/IPA + acidic conditions/time. Controls: avoid risky acid-alcohol combinations, change solvent or salt form, shorten hold time, remove alcohol before salt formation, and verify with a targeted trace-level method. C. Residual alkyl halide control Controls: drive reaction to completion, improve purge by extraction, distillation, or crystallization, and test late intermediates or API where justified. D. Degradation-driven impurity control Controls: hold-time studies, milder conditions, stability-informed strategy, and stability-indicating trace methods. Purge Factor: In the Andrew Teasdale purge factor concept, each downstream step gets purge credit based on the impurity’s reactivity, partitioning, volatility, or crystallization behavior under actual process conditions. Overall predicted purge is the product of the individual step purge factors. Example: Reaction consumption = PF 10, Aqueous extraction = PF 20, Crystallization = PF 5 PPF = 10 × 20 × 5 = 1000 If impurity level at introduction is 1500 ppm and acceptable API limit is 3 ppm: RPF = 1500 ÷ 3 = 500 Purge Margin = PPF ÷ RPF = 1000 ÷ 500 = 2 So the process is predicted to clear the impurity 2-fold more than required. But purge must be scientifically justified, not assumed. What PF > 1, = 1, or < 1 means For an individual step: - PF > 1: the step is expected to reduce the impurity - PF = 1: no meaningful purge assumed - PF < 1: the step may concentrate the impurity or give no useful clearance For the overall process: - PPF > RPF: predicted purge is adequate - PPF = RPF: just meets minimum required purge - PPF < RPF: purge is insufficient; stronger controls, more data, or direct testing may be needed M7 is not one-time paperwork. Reassess after route changes, reagent or solvent changes after the starting material, scale-up or process changes, new recycle streams, new impurity observations, formulation changes creating new degradants, or dose changes affecting exposure. Prevent formation, understand purge, test strategically, and reassess through the lifecycle. #ICHM7 #MutagenicImpurities #Nitrosamines

How can process control improve comminution efficiency?    Comminution accounts for 50-70% of a mine’s total energy consumption—more than any other process in mineral extraction.    If the process isn’t optimized, you’re literally throwing money away. Process control ensures mills and crushers run at peak efficiency, minimizing wasted energy and maximizing throughput.    Better control eliminates bottlenecks, stabilizes the process, and boosts throughput, without expensive new equipment. Just by fine-tuning mill loading, feed rates, and classification efficiency, a well-optimized system can drive a 5-15% increase in production.    What Happens Without Process Control?  Erratic feed sizes and fluctuating mill loads put extra stress on crushers, SAG mills, and ball mills, causing frequent breakdowns, shorter liner life, and rising maintenance costs.  Manual adjustments shift-to-shift, creating unstable recovery rates, fluctuating product sizes, and inefficiencies that ripple downstream.   Overgrinding wastes water, grinding media, and power—without adding value.    Step 1: Measure Everything!  The first step in optimizing comminution is knowing what’s actually happening inside the mill. You can’t control what you don’t measure. The best operations leverage real-time data on:    Mill power draw – Energy use in real time. Throughput rates – Tons per hour, ensuring consistent flow. Particle size distribution – Ensure the product meet its specification/liberation.  Cyclone performance – The right amount to circulating load avoid inefficiencies.   Step 2: Control.  Once you have real-time measurements, the next step is to stabilize the process. Better process control smooths out variability, ensuring predictable performance, higher throughput, and less energy waste. This is where operations shift from firefighting problems to running a system that self-corrects in real time.    Step 3: Optimize.  The biggest gains come when control moves beyond just reducing variation and starts pushing the process to its limits – without tipping into inefficiency. A well-optimized circuit runs leaner, faster, and more cost-effectively, reducing waste and maximizing output.    🔹 56% reduction in performance deviations 🔹 Elimination of operator bias 🔹 Higher operational efficiency and throughput   At its core, process control transforms a reactive operation into a proactive one. But there’s still one missing piece – real-time ore hardness data at the mill feed. With continuous ore characterization, operations can take process control even further, ensuring mills are operating with full knowledge of feed conditions. That’s where Geopyörä makes a difference. 

SPC (Statistical Process Control) Explained Step by Step 📊 Quality is not only about inspecting finished products — it’s about controlling the process while production is running. That’s where SPC (Statistical Process Control) becomes a powerful quality tool. SPC uses statistical methods to monitor and control process variation, helping manufacturers prevent defects before they occur. ⸻ 🔍 Step-by-Step SPC Process 1️⃣ Select the Critical Characteristic Choose the parameter that directly affects quality. Examples: 🔹 Diameter 🔹 Thickness 🔹 Flatness 🔹 Position tolerance ⸻ 2️⃣ Collect Process Data Measure samples at regular intervals during production. ✔ Use calibrated instruments ✔ Follow standard sampling frequency ✔ Record data accurately ⸻ 3️⃣ Calculate Process Average & Variation SPC mainly studies: * Process Mean (Average) * Standard Deviation (Variation) ⸻ 4️⃣ Create Control Limits Typical SPC control limits: UCL = \bar{X} + 3\sigma \qquad LCL = \bar{X} - 3\sigma Where: * UCL = Upper Control Limit * LCL = Lower Control Limit * σ = Standard Deviation ⸻ 5️⃣ Plot Control Chart Plot measurements on a control chart over time. Common SPC charts: 🔹 X̄ Chart 🔹 R Chart 🔹 P Chart 🔹 C Chart ⸻ 6️⃣ Monitor Process Stability Check for: ✔ Points outside control limits ✔ Sudden trends or shifts ✔ Abnormal patterns ✔ Process instability ⸻ 7️⃣ Take Corrective Action If abnormal variation is detected: 🔹 Identify root cause 🔹 Correct the issue 🔹 Standardize improvements ⸻ 🚀 Benefits of SPC ✅ Reduces rejection and rework ✅ Detects problems early ✅ Improves process capability ✅ Enhances customer satisfaction ✅ Supports continuous improvement ⸻ Reality Check 👇 Inspection finds defects after they happen… SPC helps prevent them before they happen. “Control the process, and quality will follow automatically.” #SPC #StatisticalProcessControl #QualityEngineering #Manufacturing #SixSigma #ProcessControl #QualityControl #ContinuousImprovement #LeanManufacturing #DataAnalysis #ProcessCapability #CpCpk #Engineering #OperationalExcellence #ZeroDefect #SPC #StatisticalProcessControl #ProcessControl #ProcessMonitoring #ProcessStability #ProcessCapability #Cp #Cpk #ControlChart #QualityTools #QualityEngineering #QualityControl #QualityManagement #QualityAssurance #ManufacturingQuality #Inspection #Metrology #DataDrivenQuality #ZeroDefect #OperationalExcellence #SixSigma #LeanManufacturing #LeanSixSigma #DMAIC #ContinuousImprovement #RootCauseAnalysis #ProblemSolving #Kaizen #WasteReduction #ProcessImprovement #Manufacturing #MechanicalEngineering #IndustrialEngineering #ProductionEngineering #Engineering #SmartManufacturing #Industry40 #Automation #FactoryOperations #ShopFloor #DataAnalysis #StatisticalAnalysis #EngineeringStatistics #Productivity #Reliability #Innovation #EngineeringLife #ManufacturingExcellence #TechnicalSkills #EngineeringCommunity

DoE, QbD and PAT 1. Introduction Evolution of pharmaceutical development: from empirical trial-and-error → risk-based scientific approaches. Regulatory drivers: ICH guidelines (Q8–Q14), FDA PAT initiative (2004). Importance of integrating design, knowledge, and real-time control. Positioning DoE, QbD, and PAT as a “triad” for robust, efficient, compliant development. 2. Historical Context and Regulatory Push Past reliance on end-product testing and its limitations. Shift to lifecycle management approaches. Role of FDA’s Critical Path Initiative. QbD introduced into regulatory lexicon in 2004; PAT guidance published. Global adoption: EMA, MHRA, WHO. 3. Understanding the Three Pillars 3.1 Quality by Design (QbD) – The Framework Definition & Philosophy: Proactive design vs reactive testing. Key Concepts: QTPP – Quality Target Product Profile. CQA – Critical Quality Attributes. CPP – Critical Process Parameters. CMA – Critical Material Attributes. Stages of Application: Early development → Technology transfer → Lifecycle management. Regulatory Basis: ICH Q8(R2), Q9, Q10, Q11, Q12, Q13, Q14. Tools: Risk assessments (FMEA, Ishikawa, Fault Tree Analysis), control strategy design. Case Study Example: QbD applied to controlled-release tablet development. 3.2 Design of Experiments (DoE) – The Optimizer Definition: Statistical framework for systematic factor–response exploration. Role in QbD: Tool to identify design space. Types of DoE: Screening designs (Plackett-Burman, Fractional Factorial). Optimization designs (Central Composite, Box-Behnken). Robustness studies. Benefits: Identifies interactions, reduces experiments, builds knowledge quantitatively. Case Example: Optimizing binder level, granulation time, and impeller speed. 3.3 Process Analytical Technology (PAT) – The Real-Time Guardian Definition: Real-time monitoring and control toolkit. Role: Ensures processes remain within validated design space. Techniques: NIR, Raman, FTIR, Particle size analyzers, Focused Beam Reflectance Measurement (FBRM). Applications: Blend uniformity. Moisture control. Coating thickness. Continuous manufacturing. Regulatory Context: FDA PAT Guidance (2004). Case Example: Inline NIR monitoring for RTRT (Real-Time Release Testing). 4. Interrelationship of the Three Pillars DoE as the engine of knowledge → defines design space. QbD as the overarching framework → integrates knowledge, risks, and control strategy. PAT as the execution safeguard → ensures adherence in manufacturing. Lifecycle integration (development → validation → continuous verification). 5. Benefits of Integrated Use Regulatory alignment & faster approvals. Cost savings through fewer failed batches. Increased robustness and reproducibility. Knowledge management & data-driven decision-making. Example: Continuous manufacturing systems where DoE defines design space, QbD integrates it, and PAT ensures execution.

How good is your process? Making a part once and making a part many times are very different tasks. We often create prototypes that lead to larger-volume production. One of the biggest challenges is being confident in your production process so you can guarantee that every part falls within your tolerance band. This is where statistical process control (SPC) comes in. At its simplest, measuring data, analyzing trends, and comparing them against upper and lower limits helps you determine whether you can trust your process. The reliability of your process can be summed up in a single number—Cp, or process capability. At NH Micro, we manufacture a lot of screws, most of them for our in-house wristwatch production. Recently, we started analyzing the dimensional data we collect to better understand our screw manufacturing process. You can see the results in the graphs below! As a quick explainer: Across 750+ parts, one dimension—a 1.90mm diameter—varied within a 6µm band. Against a ±10µm tolerance, we confirmed that our process has a Cp of 1.87 and a Cpk of 1.54. That’s really good! What’s really interesting is the trend over time. - Between 0 and 200 parts, we saw our machine warming up. - Between 200 and 675 parts, it had stabilized, but our cutting tools were slowly wearing. - At the 700th part… Well, I’ll let you guess in the comments what happened! SPC is a superpower—an incredibly useful tool for controlling your manufacturing process and pushing it to its full potential. Josh

📊 📊 SPC (Statistical Process Control) – The Complete Shop Floor Cheat Sheet Many manufacturers inspect quality only after defects occur. SPC helps you control the process before defects are produced. Instead of detecting bad parts, SPC prevents bad parts by monitoring process variation in real time. This premium cheat sheet covers: ✅ What is SPC? ✅ Common vs Special Cause Variation ✅ Control Charts (X̄-R, P, NP, C & U Charts) ✅ Process Capability (Cp, Cpk) ✅ Practical Shop Floor Example ✅ Interview Questions ✅ Common Mistakes ✅ Pro Tips ✅ Quick SPC Checklist 💡 Shop Floor Example: A hole diameter specification is 20.00 ± 0.10 mm. SPC monitoring shows a gradual upward trend before parts go out of specification. The operator adjusts the machine in time, preventing customer complaints, scrap, and rework. 🎯 Why SPC is Important: ✔ Reduces process variation ✔ Prevents defects before they occur ✔ Improves Cp & Cpk ✔ Reduces rejection and rework ✔ Improves customer satisfaction ✔ Supports IATF 16949 & AIAG requirements 💾 Save this post for future reference. 📤 Share it with your Quality Team. 💬 Interview Question: Which control chart do you use for variable data, and when would you use a P-Chart instead? Follow for more practical Quality Engineering content. #QualityEngineering #SPC #StatisticalProcessControl #QualityControl #Manufacturing #IATF16949 #AIAG #ProcessCapability #Cp #Cpk #ControlChart #LeanManufacturing #SixSigma #Production #QualityEngineer #ContinuousImprovement

SPC (Statistical Process Control) Study is a powerful quality control tool used in manufacturing and other industries to monitor, control, and improve processes through statistical methods. Here’s a detailed overview of SPC for your understanding: --- ✅ What is SPC? SPC (Statistical Process Control) is a method of using statistical tools to analyze process or production data to ensure that the process operates at its maximum potential to produce conforming products. --- 🎯 Objectives of SPC Monitor process behavior. Detect unusual variations (assignable causes). Control the process to maintain consistency. Improve process capability and reduce defects. --- 🧰 Key SPC Tools 1. Control Charts (Most Important) X̄ (X-bar) and R Chart – for variable data (mean and range). X̄ and S Chart – for mean and standard deviation. p Chart – for proportion defective. np Chart – for number of defectives. c Chart – for count of defects. u Chart – for defects per unit. 2. Histogram Shows frequency distribution of data. 3. Pareto Chart Highlights major causes of defects using the 80/20 rule. 4. Cause-and-Effect Diagram (Fishbone/Ishikawa) Identifies root causes of variations. 5. Scatter Diagram Shows correlation between two variables. --- 🔍 Types of Variation 1. Common Cause Variation Natural variation in the process. Cannot be easily eliminated. 2. Special Cause Variation Arises due to specific issues (machine failure, human error). Should be investigated and eliminated. --- 🧪 Steps of an SPC Study 1. Select the process to be studied. 2. Identify key characteristics (CTQs – Critical to Quality). 3. Collect data over time. 4. Create control charts. 5. Interpret the charts (look for trends, out-of-control points). 6. Take corrective action if needed. 7. Maintain control by continuous monitoring. --- 📏 Control Limits vs Specification Limits Control Limits (UCL, LCL): Based on process data, reflect process stability. Specification Limits (USL, LSL): Based on customer requirements. > A process can be in control (within UCL/LCL) but out of specification (outside USL/LSL) and vice versa. --- 📈 Benefits of SPC Reduced variability and scrap. Improved product quality. Better process understanding. Predictable process performance. Lower production costs. Vivek Pandey

SPC is more than just UCL and LCL. The 8 Nelson Rules help spot unusual patterns within control limits that show your process might be unstable. Catch these early to avoid problems. The 8 Nelson Rules: Rule 1: One point > 3 standard deviations from mean. Rule 2 : Nine points in a row on the same side of mean. Rule 3 : Six points increasing or decreasing in a row. Rule 4: Fourteen points alternating up and down. Rule 5: Two of three points > 2 standard deviations (same side). Rule 6: Four of five points > 1 standard deviation (same side). Rule 7 : Fifteen points in a row within 1 standard deviation (too consistent). Rule 8: Eight points in a row outside 1 standard deviation (both sides). #StatisticalProcessControl #SPC #QualityControl #QualityAssurance #QualityManagement #SixSigma #ProcessImprovement #DataAnalytics #LeanManufacturing #Manufacturing #ContinuousImprovement

𝐈𝐟 𝐲𝐨𝐮 𝐭𝐡𝐢𝐧𝐤 𝐯𝐚𝐥𝐯𝐞𝐬 𝐬𝐢𝐦𝐩𝐥𝐲 “𝐨𝐩𝐞𝐧 𝐚𝐧𝐝 𝐜𝐥𝐨𝐬𝐞,” 𝐲𝐨𝐮’𝐫𝐞 𝐦𝐢𝐬𝐬𝐢𝐧𝐠 𝐡𝐨𝐰 𝐩𝐫𝐨𝐜𝐞𝐬𝐬 𝐜𝐨𝐧𝐭𝐫𝐨𝐥 𝐚𝐜𝐭𝐮𝐚𝐥𝐥𝐲 𝐰𝐨𝐫𝐤𝐬. Flow in a pipe depends mainly on: - Pressure difference - Flow area - Flow resistance A valve controls flow by varying the effective flow area, which changes resistance. Larger opening → lower resistance → higher flow Smaller opening → higher resistance → lower flow This relationship is governed by Bernoulli’s equation and valve flow coefficients (Cv/Kv). How does flow regulation happens actually? Unlike manual valves, a control valve is a dynamic device. It does not stay fixed, it continuously responds to process conditions to maintain the desired flow. 1. This happens through a closed-loop control system. 📍 A flow transmitter (FT) measures the actual flow rate using a Orifice plate Venturimeter or Magnetic flowmeter 📍The transmitter converts the measured flow into a standard signal: 4–20 mA (most common) or digital (Fieldbus, Profibus) 📌 This signal represents the current state of the process, not what we want. Example: Setpoint = 100 m³/h Measured flow = 92 m³/h The process has spoken: “I’m below target.” 2. The signal goes to a controller (PLC or DCS). The controller compares: Setpoint (SP) – desired flow Process Variable (PV) – actual flow The difference is called error: Error = SP − PV Using a PID algorithm: - P (Proportional): reacts to current error - I (Integral): removes steady-state offset - D (Derivative): anticipates future error The controller calculates how much the valve must move. 📌 This is why tuning matters. Poor PID tuning = oscillating flow, hunting valves, unstable operation. 3. Control signal to actuator 📍The controller sends an output signal to the valve: 4–20 mA → I/P converter → 3–15 psi air Or direct electric signal for motorized actuators This signal tells the actuator: “Open more” or “Close slightly” 📍The actuator provides the force required to overcome: Process pressure Fluid forces Packing friction Spring force (fail-safe action) 4. Valve movement - where flow actually changes Now comes the physical control. Inside the control valve: 📍The plug, ball, or disc moves relative to the seat 📍The effective flow area changes 📍Flow resistance changes 📍Pressure drop across the valve adjusts. This directly alters the flow rate. 📌 Important insight: A control valve does not control flow directly. It controls pressure drop, and flow responds to that pressure drop. 5. Feedback and stabilization: the loop closes Once the valve moves: Flow changes Transmitter senses the new flow New signal goes back to the controller This repeats every second (or faster) until: PV ≈ SP At this point, the system is stable. This continuous correction is why it’s called a closed-loop control system.