Comment interprétez-vous les valeurs Cp et Cpk par rapport à la capacité et à la variation du processus ?

The statement that both Cp and Cpk are calculated by dividing the difference between the USL and LSL by six times the standard deviation is incorrect. For Cp, the formula is indeed (USL - LSL) / (6 * σ). However, for Cpk, the calculation involves determining the distance from the mean to the nearest specification limit, divided by three times the standard deviation: Cpk = min[(USL - μ) / (3 * σ), (μ - LSL) / (3 * σ)].

Cp and Cpk are invaluable tools for unlocking process capability. While Cp highlights potential performance, Cpk provides a realistic measure of actual capability. Together, they empower organizations to optimize processes, reduce variation, and ensure consistent quality. Cp, or Process Capability, measures the potential capability of a process to meet specification limits. It evaluates the ratio of the specification range (USL - LSL) to the natural process variation (6σ), assuming the process is perfectly centered. Cpk, or Process Capability Index , measures the actual capability of a process. It accounts for both variation and how well the process is centered within the specification limits.

In my experience, during a project to improve the quality of hard disk components, we calculated Cp and Cpk to evaluate the manufacturing process. Initially, a low Cpk indicated that the process was not well-centered, resulting in defects. By realigning the process parameters and reducing variability, we improved the Cpk, ensuring a higher percentage of components met the quality standards.

Cp and Cpk values come only if the process is stable, otherwise they will be Pp and Ppk. Cp is the value that gives what the process capability would be if the process was at the center. What is meant by the center is determined by the distance of the sigma value from the average value. While the sigma value is the average of the results of current measurements, the average value indicates the required value. It shows the current process capability in Cpk.

Cp is potential process capability which shows how much spreaded process is and Cpk is actual process capability which tells us shift in the process. Cpk cannot be higher than Cp. In short Cp is spread and Cpk is shift.

DPMO: Your Quality Scorecard DPMO stands for Defects Per Million Opportunities. It's like counting how many mistakes you make for every million chances. A low DPMO means you're doing a great job, while a high one signals a problem. Imagine a factory making toys: if 10 out of 10,000 toys have defects, the DPMO is 1,000. The goal is to get that number as close to zero as possible.It's a simple yet powerful way to measure quality and track improvement over time.

A complex unit with multiple components will have more defect opportunities than a simple one. It's important to differentiate between defects (individual flaws) and defectives (units with one or more flaws). In DPMO, we focus on the count of defects, not the number of defective units.

Brindo otro ejemplo: Cuando se diligencia a mano un formulario con 40 campos, usted tiene 40 oportunidades para fallar y pudo haber tenido 1 error. 1/40 = 0.025 x 1.000.000, dpmo = 25.000

O valor de Cp é um indicador que compara a quantidade de variação entregue pelo processo com a quantidade de variação especificada. O Cpk traz o resultado final da capabilidade, levando em conta a variação e o deslocamento da média dos dados capturados vs a especificação. Ambos são essenciais para direcionar ações, tendo em mente o aumento de performance ou a busca para se atender o que foi determinado pela voz do cliente.

Utilize both Cp and Cpk to pinpoint improvement opportunities. A high Cp coupled with a low Cpk indicates the process needs centering. Focus on reducing variation and aligning the process mean to the target specification. Benefits: 1. Enhanced Performance: Higher Cp and Cpk values signify superior process performance and stability. 2. Informed Decisions: By analyzing both values together, you gain a comprehensive understanding of your process capability and stability. 3. Targeted Improvements: Addressing discrepancies between Cp and Cpk ensures targeted, effective process enhancements. Higher values typically denote better performance, but always consider the relationship between Cp, Cpk, and the process target for a thorough analysis.

O indicador Cp só leva em consideração a tolerância do cliente em relação à variação interna do processo (VOC/VOB): (LSE-LIE)/(6 x desvio padrão) e não considera o deslocamento de uma média. A análise é sobre a largura total da curva. Já o Cpk leva em consideração um possível deslocamento de uma média do processo dentro da tolerância do cliente: Ele analisa separadamente os dois lados da curva de distribuição: (LSE-X)/(3 x desvio padrão) para analisar o lado direito da curva e (X-LIE)/(3 x desvio padrão) para analisar o lado esquerdo da curva. O Cpk a ser considerado será o de menor índice dos dois.

The best way to interpret these values is below equation : 3*Cpk= Actual sigma of process. For example, if Cpk is 2 then 3*2= 6 sigma process. Cpk is 1.67 then 3*1.67= 5 sigma process.

The Cp & Cpk values measure the consistency of the process as compared to its average performance. The classic " Parking a Car in the Garage" analogy explains the concept in simple terms. To ensure that the car is parked perfectly, we should reduce the variations (in this case, the width of garage or size of the car) & maintain central alignment during parking. Cpk indicates the relationship between the car's size, the garage's size, and how far the car is parked from the center of the garage. Higher Cpk denotes that the process is less likely to go out of specifications.

In an electronics assembly process, a Cp value of 1.4 indicated that the process was capable of staying within specification limits. However, Cpk was 1.0, which highlighted a process shift closer to one specification boundary. When I calculated the DPMO, it revealed around 3,000 defects per million opportunities, far more than expected based on Cp alone. This difference between Cp and Cpk showed that focusing solely on Cp can give a false sense of process stability, and that aligning the process more centrally (as Cpk reflects) can drastically reduce defects.

A relação do Cpk com DPMO está relacionada ao sinal do indicador. Cpk negativo = O processo está com a média fora de uma das especificações do cliente entregando mais de 500 mil defeitos por milhão de oportunidades. Cpk zerado = O processo está com a média sobreposta em uma das especificações do cliente entregando 500 mil defeitos por milhão de oportunidades. Cpk positivo = O processo está com a média dentro das especificações do cliente entregando menos de 500 mil defeitos por milhão de oportunidades.

Em geral, se busca um aumento de performance através do estudo das relações de causa e efeito do processo. Isso pode ser realizado através de diagramas de Ishikawa, técnica dos 5 porquês e/ou análises de variáveis de entrada e saída como em um COV ou em um DOE, que são recursos que exigem um conhecimento prévio de estatística. Para um nível reduzido de variação (Cp alto) e um deslocamento de média, resultando em um Cpk baixo, as tratativas são mais simples e dificilmente requerem grandes investimentos de tempo ou dinheiro. Para os casos de uma grande variação (Cp e Cpk baixos), a definição de projetos e os métodos são mais estruturados, levando a coletas de dados e consequentemente será necessário mais tempo para se chegar à solução final.

Quoting my real life experience, in a hard disk component manufacturing project, I observed that high variability in machine calibration was affecting our Cp and Cpk values. By standardizing calibration procedures and conducting regular maintenance, we significantly reduced process variation. Additionally, training operators on best practices and using SPC charts to monitor the process helped us maintain a centered process, resulting in improved Cp and Cpk values and higher product quality.

The best way is to identify source of variation, do standardisation in processes, training and education, close feedback loop.

Improving Cp and Cpk values is critical for achieving high-quality, consistent processes. By reducing variation, centering the mean, enhancing input quality, leveraging technology, and fostering a culture of continuous improvement, organizations can significantly enhance their process capability. This, in turn, leads to reduced defect rates, improved customer satisfaction, and a stronger competitive edge. Remember, process improvement is a journey, not a destination—continuous efforts will yield long-term rewards.