Descriptive Analysis
Descriptive analytics examine what has happened in history to identify patterns in data. These insights can come from internal supply chain execution software as well as from external systems that provide visibility into suppliers, distributors, different sales channels and customers.
Data driven Analytics can compare the same type of data in real time over different time frames to pinpoint patterns and generate hypotheses about the potential causes of changes.
An electronical device company can look at a descriptive analytics dashboard on a daily basis and discover that half of its deliveries to distributors are late. The company's executives can then dig deeper into this problem and learn that trucks have slowed down due to a torrential rain in the district around this group of distributors.
Predictive analytics
As the name suggests, predictive analytics takes the guesswork out of predicting what might happen and the business impact of various scenarios, from potential supply chain disturbances to other outputs. By forcing executives to consider these possible scenarios before they happen, they are enabled to act proactively rather than reactively. It gives them time to develop a strategy for dealing with an expected increase or decrease in demand, for example, and they can respond correspondingly.
Considering the same electronical device company, it may examine the latest economic projections from the Federal Reserve and anticipate a 10 to 20 percent decline in sales in the next quarter. In light of this, he orders fewer raw materials from his providers and lowers the working hours of part-time staff for the following 30 days. It could therefore be a real competitive advantage if competitors do not have the same metrics and perspectives.
Prescriptive Analysis
Prescriptive analysis combines the results of descriptive and predictive analysis to recommend actions a company should take now to accomplish its goals. Such analysis could help companies fix challenges and avoid major supply chain disruptions, potentially by evaluating both its own information and its partners' information. Since prescriptive analytics are more complex, they require more powerful software that can quickly process and interpret massive amounts of data.
Prescriptive analytics can tell our electronical device company that one of its key suppliers in Western Europe is likely to go out of business in the next year. A consistent history of back orders, reduced capacity, and declining economic conditions in the region are all indicators of this risk.
In response, the manager could request a meeting with the supplier's management to determine if they are in financial trouble and how it could help. If no clear solution is found, the company can begin to look for other suppliers to replace this one before it's too late.
Cognitive Analytics
Cognitive analytics attempt to replicate human thinking and behavior to help companies answer difficult and complex questions. These analytics are able to understand real precise information like context when interpreting results. To do so, cognitive analytics relies on artificial intelligence (AI), including machine learning and deep learning, which allows it to become more intelligent over time. This can significantly reduce the volume of work needed, of people to produce that reporting and analysis, and can grant a great degree of autonomy to people beyond the data science squad to draw out the results and figure them out.
With its Artificial Intelligence-powered software, our electronics company could be able to further automate a significant amount of demand planning work. The system could process all existing data, as well as both internal and external factors, to make very accurate and detailed suggestions on how much of each product it should manufacture for the upcoming period to meet demand. This would cut down on the extra expense of building up more stock than needed or lost revenue due to the inability to satisfy demand.
