• Written by Sonali Subhadarsini, Product Manager

A shift supervisor on the night shift in a chemical plant often has to wait for the morning report to understand how the previous day's production performed against plan, and where the bottlenecks were, in order to adjust the production schedule for the current day. This wait can be frustrating, especially when the production process is continuous and any disruption can have a significant impact on yield and safety. The supervisor may have to rely on manual data collection and phone calls to various teams to get a sense of what is happening on the plant floor.

The production questions that currently wait on a report

The production team in a chemical plant typically has a set of questions that they need to answer on a daily basis, such as what was the actual production volume against the planned volume, what were the cycle times for each batch, and what was the quality of the output. These questions are critical to understanding how the production process is performing and where improvements can be made. However, the answers to these questions often have to wait for the daily or weekly report to be generated, which can be time-consuming and may not provide the level of detail that the team needs.

Most people get wrong the idea that these reports are sufficient for the production team to make informed decisions. They are wrong because the reports are often delayed, and the data is not granular enough to allow for real-time decision-making. The production team needs to be able to ask questions of their data in real-time, such as what is the current production volume, or what is the status of the maintenance schedule, in order to make adjustments on the fly.

The data behind them: output against plan, cycle times, scrap and quality records, maintenance history

The data that the production team needs to answer their questions is typically stored in various systems, such as the enterprise resource planning (ERP) system, the manufacturing execution system (MES), and the computerized maintenance management system (CMMS). This data includes output against plan, cycle times, scrap and quality records, and maintenance history. By having access to this data in real-time, the production team can get a complete picture of the production process and make informed decisions.

For example, a category manager can compare the production volume of different products and adjust the production schedule accordingly. The category manager can also use the data to identify trends and patterns in the production process, such as which products are more prone to scrap or quality issues, and adjust the production schedule to minimize these issues.

Sector constraints: yield drift and feedstock variability

The chemical industry has its own set of constraints that can impact production, such as yield drift and feedstock variability. Yield drift refers to the gradual decline in production volume over time due to equipment wear and tear, while feedstock variability refers to the changes in the quality and composition of the raw materials used in production. These constraints can have a significant impact on production and need to be taken into account when making decisions.

The production team needs to be able to ask questions of their data that take into account these constraints, such as what is the current yield drift for a particular product, or what is the impact of feedstock variability on production volume. By having access to this data in real-time, the production team can make adjustments to the production schedule to minimize the impact of these constraints.

What improvement looks like: OEE, unplanned downtime hours, scrap rate

Improvement in the chemical industry can be measured in various ways, such as overall equipment effectiveness (OEE), unplanned downtime hours, and scrap rate. OEE measures the percentage of time that equipment is operating at its maximum capacity, while unplanned downtime hours measure the amount of time that equipment is down due to unexpected maintenance or other issues. Scrap rate measures the percentage of defective products that are produced.

By using AI for production in chemicals, the production team can get a real-time view of these metrics and make adjustments to the production schedule to improve them. For example, the production team can use AI-powered analytics to identify the root causes of unplanned downtime and take steps to prevent them. The team can also use AI to optimize the production schedule to minimize scrap rate and improve OEE.

What this does not do, or when it is the wrong choice

While AI for production in chemicals can be a powerful tool for improving production, it is not a replacement for human judgment and expertise. The production team still needs to use their knowledge and experience to interpret the data and make decisions. Additionally, AI for production in chemicals is not suitable for all types of production processes, such as those that require a high degree of manual intervention or those that have a high level of variability.

If you are considering using AI for production in chemicals, it is recommended that you contact us to discuss your specific needs and determine whether AI is the right choice for your organization. We can help you understand how AI can be used to improve your production process and provide you with the support you need to get started.