Why we need high quality data
Supply Chain analytics relies heavily on optimization algorithms that need high quality data to generate accurate models. To correctly improve operations, the data used in businesses models needs to be relied on enough to make important business decisions. If you are unsure about the quality of your data review the potential data errors below.
Data collection errors
- Duplication of data
- Non-conforming (data that ignores constraints)
- Inconsistant data
- Inaccessible data
- Incomplete data
Data errors in the research or survey process
- Population Specification – selection of an inappropriate universe to obtain data
- Sampling – sample is not an accurate representation of the population
- Selection – collecting data from where is it most accessible
- Non-responsive – occurs when sample differs from the original selected sample
- Measurement -difference of information generated and information wanted
What Goes Wrong With Unhealthy Data?
Simple guessing estimates for increase productivity can lead to major impacts to a companies financial statements. A business’s operating margin is often sensitive to productivity and efficiencies calculated with historical data. Manufactures risk severe complications if their data represents a different picture than the reality of the situation. This could lead to large purchases of unnecessary equipment, a overpaid and under utilized staff, and high inventory levels leading to stopping production.