Steven Wachs

Vice President & Principal Statistician, Integral Concepts, Inc.

Steven Wachs has 25 years of wide-ranging industry experience in both technical and management positions. He has worked as a statistician at Ford Motor Company where he has extensive experience in the development of statistical models, reliability analysis, designed experimentation, and statistical process control.

Mr. Wachs is currently a Principal Statistician at Integral Concepts, Inc. where he assists manufacturers in the application of statistical methods to reduce variation and improve quality and productivity. He also possesses expertise in the application of reliability methods to achieve robust and reliable products as well as estimate and reduce warranty. Mr. Wachs regularly speaks at industry conferences and provides workshops in industrial statistical methods worldwide. 

He has an M.A. in Applied Statistics from the University of Michigan, an M.B.A, Katz Graduate School of Business from the University of Pittsburgh, 1992, and a B.S., Mechanical Engineering from the University of Michigan.

Recorded-webinar by: Steven Wachs

    • 90 mins

      Crucial Role of Sample Sizes in Design Validation: Evaluating Product Performance, Quality, and Reliability

      Life Sciences

      The process of design validation is a crucial step in ensuring the quality and effectiveness of a product. However, determining the appropriate sample size for design validation activities can be challenging. A sample size that is too small may not provide sufficient data to make accurate conclusions, while a sample size that is too large can be costly and time-consuming. 

      Statistical Methods are typically used to ensure that product performance, quality, and reliability requirements are met during the Design Validation phase of product development. 

      This webinar discusses common elements of sample size determination and several specific sample size applications for various design validation activities including Reliability Demonstration/Estimation, Acceptance Sampling, and Hypothesis Testing. By the end of the webinar, participants will have a better understanding of how to determine the appropriate sample size for design validation activities, and how to ensure that the sample size selected is sufficient to make accurate conclusions.

    • 75 mins

      Ensuring Consistent Quality: Effective Methods for Assessing Process Capability for Normal and Non-Normal Data

      Life Sciences

      To ensure consistent production of products and services that meet customer specifications, companies must ensure their processes are capable.  

      This webinar covers techniques for estimating process capability for both normal and non-normal data. The session first discusses pre-requisites for estimating process capability, such as establishing process stability. It then briefly describes distributions and presents methods for estimating ppm levels. The webinar also delves into the use and limitations of common process capability indices like Cpk and Ppk.

      Inaccurate estimation of capability due to inappropriate methods for non-normal data can lead to over-optimistic results, making it crucial to use suitable techniques. The webinar discusses methods for testing normality and presents transformations and distribution fitting as effective approaches for assessing non-normal data. It also includes several examples to explain the methods in detail.

    • 90 mins

      Acceptance Sampling Plans for Process Validation and Production Lot Monitoring

      Life Sciences

      This webinar covers Acceptance Sampling plans for process validation and production lot acceptance. 

      Sampling plans for attribute data are the primary focus although variable acceptance sampling plans are presented as well. The binomial distribution and its use in developing Operating Characteristic (OC) Curves is discussed. The key inputs to determining sampling plans (AQL, RQL, Consumer's and Producer's Risks) are described in detail. Key characteristics of the generated sampling plans (such as average outgoing quality) are presented. Double sampling plans are briefly introduced. Several example applications of acceptance sampling are presented. The use of Statistical Process Control and Process Capability methods are presented as an alternative to variable acceptance sampling plans.

    • 75 mins

      Estimating and Demonstrating Product Reliability: Achieving Optimal Performance

      Engineering & Manufacturing

      Product Reliability requirements must be satisfied like all other performance requirements prior to product launch.  However, since reliability is a function of time, the methods for verifying that reliability performance has been verified differ from most other performance characteristics.  This webinar will present several approaches for verifying that reliability targets or specifications have been achieved at the desired level of confidence.  Specifically, approaches using time-to-failure data to estimate reliability metrics as well as demonstration tests, where minimum reliability may be demonstrated with zero or few failures are discussed. 

      This webinar provides methods that allow the risks of field failures due to inadequate designs or misunderstanding of product use conditions to be managed.  Also, the webinar provides options to verify and demonstrate that customer reliability requirements have been achieved.

    • 90 mins

      Stability Studies and Estimating Shelf Life with Regression Models

      Life Sciences

      Manufacturers of foods, drugs, consumer goods, and other products must determine the shelf life of their products so that customers know when the product can be expected to perform as intended. Many approaches are available to quantify the "shelf life" and the method(s) chosen often depend on the testing time available.  

      In this webinar expert speaker Steven Wachs, will clearly explain the steps to set-up a stability study and analyze the results to estimate the product's shelf life. The use of regression models to model the relationship between the response variable(s) and time are presented. Models useful for describing non-linear degradation over time are also presented. Additionally, methods for handling non-normal response data are also discussed. Finally, the use of accelerating variables to shorten the study time and the models required are introduced. 

      The webinar includes several examples to illustrate the methods discussed.