1. Introduction

Working with manufacturers around the globe, Sight Machine has seen that the success of digital manufacturing analytic projects depends on the fitness or availability of relevant data and the readiness or usability of that data. This document provides guidance on how to ensure that the appropriate data is both available and usable to ensure successful manufacturing analytic projects.

A useful metaphor for understanding the data requirements for analytics projects is that of baking a recipe. Data fitness is having the ingredients outlined by the recipe, while data readiness is ensuring that the ingredients are ready for use (cleaned, chopped, or ground to the requirements of the recipe).

  • Data fitness, which is discussed here in detail, is the ability of the data to support a use case or set of use cases. For example, in defect analysis, it is necessary to have quality data and process data to determine what may be contributing to the defective material. It is also necessary to have a method for linking the data sources (such as serial numbers) together.
  • Data readiness, the ability to use one’s data, is the first step in ensuring that the data is in a state that is accessible and useable. For example, data is not accessible if it is in programmable logic controllers (PLCs), in a proprietary format, or contains out-of-sync time stamps.

The remainder of this document does the following:

  • Outlines the approach for surveying the manufacturing data landscape.
  • Discusses key points of data readiness.
  • Provides a high-level mapping between use cases and data sources.
  • Discusses the details of use cases supported by the Sight Machine platform.
  • Describes manufacturing data sources utilized for the enablement of supported use cases.