Incident Management
From Investigation to Management: The Danone Approach
In modern industry, incident investigation is only a small part of risk management. It is much more important to build an incident management system that will prevent them in the future. In his presentation, Dmitry Zubov, HSE Director at Danone, shares practical experience and explains how the company transitioned from simply looking for culprits to a deep analysis of root causes and risk forecasting.
Safety Principles: From Zero Injuries to Personal Responsibility
The speaker emphasizes that Danone's safety system is based on several key principles. First, the company believes that all accidents can be prevented, and the only acceptable goal is zero incidents. Second, safety depends on the behavior of every employee. Third, the primary responsibility for safety lies with line managers. Finally, no profit justifies violating safety rules.
Investigating All Incidents: Why It Matters
At Danone, absolutely all incidents are investigated, including micro-injuries, first aid cases, and near-misses. This is due to the large number of outsourced personnel on site. The company involves contractors in its safety system on the same basis as full-time employees. This approach allows identifying systemic problems and preventing more serious incidents.
Investigation Methodology: A Combination of Tools
To investigate incidents at Danone, a combination of methods is used: the Ishikawa diagram (5M/6M) and the "5 Whys" analysis. The speaker analyzes common mistakes when using these methods separately. For example, the "5 Whys" analysis without 5M may not reveal all the causes of an incident, and using only 5M may not lead to the root causes. The combination of methods provides an objective picture and helps develop an effective corrective action plan.
Incident Forecasting: A Look into the Future
The presentation details the transition from a reactive to a proactive approach. Danone is developing an incident forecasting system using neural networks. The system will analyze a huge array of data: employee profiles, risk levels by location, audit results, business situations, and other factors. This will make it possible to identify "warning signs" and prevent incidents before they occur.
What you will learn from this webinar:
- How to build an effective incident reporting system?
- Why is it important to investigate micro-injuries and near-miss incidents?
- How to properly combine 5M and "5 Whys" methods to find root causes?
- How to evaluate the effectiveness of corrective actions?
- What data can be used to forecast incidents using neural networks?