The previous post discusses two of the four stages in Smart Factory evolution. The first step deals with data availability. In this stage, the manufacturers cannot do much, since the data are available in separate places. In the second stage, the data becomes accessible, since the manufacturers invest infrastructure to integrate the data into a single database system. Limitation of the stage two includes inability to anticipate the problems before they really happen.
Stages of Smart Factory Evolution: #3 Active Data
In the stage two, the manufacturers can take immediate actions to address the problems by using a single database. However, they cannot take preventive actions by using the database. In the stage #3, the operation shifts from reactive problem solving to proactive one. The manufacturers can do proactive analysis and improvements by using the data.
So, what do the manufacturers need to move from stage #2 to stage #3? They include adding a new system capabilities into the system built in the stage #2. Examples include machine learning and Artificial Intelligence (AI). These tools help the manufacturers in many ways, including:
- Generating insights on the data systems in 2 or 3 months;
- Aggregating all production data into a single system;
- Predicting failures, like product problems and machine failures, in a more accurate way;
- Delivering information at the right time;
- Eliminating the need to build a query system in a manual process analysis;
- Identifies strategies for more efficient productions
However, the stage #3 still has certain limitations. For instance, it is not fully automatic. Human resources are still important to make necessary changes as required by the intelligent system.
Stages of Smart Factory Evolution: #4 Action-Oriented Data
This is the final stage of Smart Factory Evolution. In this stage, the intelligent system is able to provide recommendations based on the analysis on manufacturing data. For instance, when the machine learning identifies problem, it then generates and sends recommendations. It may recommend new settings to the machine in a fully automatic way.
When the manufacturers reach this stage, the production line becomes fully controlled by Artificial Intelligence. Two achieve the stage #4, the company needs to have data sets with enough capacity and enough validated cases. They are important to provide the information required to identify the impacts of a production change. Moving from level #3 to level #4 requires different time, depending of the time necessary to build the data sets.
Stages of Smart Factory Evolution: The Bottom Line
There are four stages for a manufacturer to build a Smart Factory. This shows that there are no shortcuts to building a factory, which is fully support by intelligent systems. In addition, the journey of smart factory evolution varies widely, depending upon the existing conditions of the manufacturer. Therefore, the investment necessary also varies.
In conclusion, the manufacturers need to make step-by-step approach to get thorough the evolution stages. They need to identify where they are now, and then plan the strategies for each stage. Step-by-step approaches increases the success rate, reduces the time to build the system, and reduces the stress during the process.