AI-based personnel planning: The invisible helper for the warehouse. (Part 1/3)

As we know, artificial intelligence (AI) can be used to automate and optimize warehouse processes. This is the case with our "Warehouse Healing" strategy, which shortens travel times, improves warehouse structures and reduces throughput times during order picking.

But to fully exploit the potential for increasing efficiency in the warehouse, our data science expert Timofej is working on AI-based personnel planning for intralogistics.

Melanie from Marketing spoke to Timofej about the topic and learned some exciting things.

1. why did you get involved with AI-based personnel planning?

Timofej: As a data scientist at a software manufacturer for warehouse management systems (WMS), I am always looking for ways to improve and automate processes. The idea of using artificial intelligence (AI) for workforce planning came about while I was assisting with a bachelor's thesis. We analyzed the data from our WMS, including past incoming orders, stock levels and historical data, and discovered the potential for optimizing workflows.

In logistics, the unpredictability of the volume of work often presents a challenge. Scheduling the right number of employees at the right time is not easy. AI-based personnel planning recognizes patterns and predicts peak phases. This enables more efficient planning. The aim is to plan and act not only reactively, but also proactively.

2 I often read on the Internet: "We use AI for personnel planning." Is that really the case?

Timofej: The claim "We use AI for HR planning" is quite broad and is interpreted in different ways. Many companies say they use AI, but in reality they only use simple automation techniques or statistical methods. These are often found in tools such as Excel or specialized software. While these methods are useful, they do not offer the advanced capabilities of true AI models.

The development of AI-based workforce planning goes beyond simple automation and traditional data analysis. We develop and implement advanced AI models that learn from large amounts of data. These models recognize complex patterns and make precise predictions for workforce planning.

The AI-based forecasts in our WMS help users to make dynamic and data-driven decisions. Integration takes place directly in the operating environment.

3. how can AI support personnel planning?

Timofej: Artificial intelligence and machine learning in personnel planning offer various opportunities to optimize processes and increase efficiency.

1. Predicting the workload
2. Optimization of shift schedules and resource allocation
3. Data-driven decision-making
4. Analysis and recommendations for action
5. Personnel development and training
6. Improving employee satisfaction

4. which problems can be solved with the help of AI in personnel planning?

Timofej: Manual personnel planning is a time-consuming task and very error-prone in a complex and large warehouse with many employees and requirements. These problems can now be eliminated:

1. Unpredictability of demand
2. Incorrect manual planning
3. Over- and understaffing
4. Lack of flexibility
5. Inefficiency and high costs
   

5. what data is required?

Timofej: For effective and precise AI-based personnel planning, I need a broad database. The following data is important:

1. Historical order data from the WMS
2. Stocks from the WMS
3. Movement data from the WMS
4. Employee data on qualifications and availability (anonymized, if necessary)
5. Historical shift schedules and timesheets (if available)
6. Seasonal trends and external factors such as public holidays

6. what challenges are there during implementation and use?

Timofej:

1. Data quality and availability pose major challenges. Incomplete data impairs the performance of the AI and outdated data distorts the results. That's why I have built in an additional data check and cleansing function.
2. The integration of AI-based personnel planning into existing systems is complex. This requires seamless interfaces for data exchange.
3. Employee concerns that AI could replace workers are another challenge. Transparent communication and training are necessary to create acceptance.
4. The modeling of AI for personnel planning purposes is complex. Many variables have to be taken into account and constant adjustments to changing conditions are necessary.
5. We must comply with the legal and ethical framework, particularly with regard to data protection and employee data. Continuous communication with employees is important.
6. The AI model must be scalable in order to adapt to company growth and changes.

7 I'm getting ahead of myself, as the tool is still in the development phase. How will the tool be made available in the WMS and will there be a separate dashboard?

Timofej: Yes, that's the plan. The forecasts for staffing requirements and other analyses are then integrated directly into the WMS. This guarantees a seamless user experience. Customers can access all information as usual. The dashboard is designed to be clear and user-friendly. It offers diagrams, tables and visual representations. These help users to analyze and interpret data effectively.    

The dashboard provides continuous updates so that specialist staff can act quickly. The forecasts in the WMS ensure that all relevant information is centralized and easily accessible. This reduces administrative work and speeds up decision-making. The integration minimizes the need for training, as employees can continue to use their familiar systems.

Thank you for your time and the interesting insight into your work.

‍

In Part 2, we explain the advantages and possibilities in more detail: AI-based workforce planning: Optimum scheduling with just a few clicks (Part 2/3).

In Part 3, we explain how our data science expert goes about developing the AI model for workforce planning: AI-based workforce planning: Creating added value from data (Part 3/3).

We use data from our customer Hermann Müller Elektrogrosshandel GmbH to develop the tool. You can find out more in this article.