How an Automotive Manufacturer Improved Line-Side Parts Delivery With a Route Planning Algorithm

A plant layout redesign changed travel routes, workstation locations, and the time needed to deliver parts. The existing algorithm continued assigning tasks to operators, but it did not account for the new conditions. Components arrived too late, putting the production line at risk of downtime. Another goal of the optimization project was to reduce the number of vehicles used to deliver materials to the assembly line.

The explitia team expanded the system to include actual travel distances, new traffic rules, and an analysis of all available route options. Operators see their next tasks on a mobile device, while the route plan changes as work progresses.

The system managed the flow of parts between the warehouse and production lines. It told the operator which component to collect and which workstation to deliver it to. It also reserved the order so that it would not be assigned to two people.

After the plant layout redesign, routes, distances, and traffic organization changed. New equipment affected how operators could move through the facility, and some workstations could only be accessed from a specific direction.

The previous algorithm still reserved parts correctly, but it did not estimate the time required for delivery. It selected from five predefined routes, even though the new plant layout required the analysis of many more options.

In automotive manufacturing, a delay in one process can affect other areas of the plant. Even several minutes of downtime can result in measurable financial losses. A production stoppage lasting several or a dozen minutes may lead to losses of tens of thousands of euros.

The new algorithm checks workstation demand and estimates when components will run out. When inventory approaches its lower limit, the operator receives a task early enough to collect the parts and travel to the workstation.

Priority depends on the current situation on the production floor. The system evaluates which delivery is most urgent and whether the operator can complete it on time.

The operator sees what to transport, where to go, and which steps to complete along the route. After each step is confirmed, the route is recalculated and the application displays the next task.

A distance map of the plant was added to the system. As a result, the algorithm can assess thousands of combinations instead of choosing from a small number of predefined options.

Three main parameters affect the decision:

Each combination receives a score based on these data points. The system selects a route that completes urgent deliveries while limiting unnecessary travel.

It also accounts for one-way and two-way traffic, as well as the permitted side of approach to each workstation. The selected route therefore reflects the actual plant layout.

The solution runs on mobile devices installed on forklifts operated by employees. The customer did not have to replace the entire fleet with autonomous vehicles or redesign material transport around an AGV system.

The method of planning and assigning tasks changed. Operators continue using familiar equipment, but the order of trips is based on current data.

Such a solution is especially relevant in older plants, where full transport automation would require a much larger investment.

An incorrect system setting could delay a delivery and stop the production line. The team therefore needed access to data showing how the algorithm made each decision.

We added event logging and operator movement records. We also created an analytics module that displays routes and compares available options.

The customer’s team can check:

The data helps set the right balance between delivery time, travel distance, and the number of orders completed during one trip.

The system predicts when line-side inventory will run out and includes the time the operator needs to reach the workstation. Urgent tasks move higher in the queue before a parts shortage affects production.

Excess parts take up space, block travel routes, and make changeovers more difficult. After a production variant changes, unused components must be returned to the warehouse.

The algorithm supports just-in-time delivery, so the quantity delivered to the line matches the current production plan.

During one trip, an operator may complete several orders, provided that doing so does not delay a more urgent delivery. The order of tasks is based on current data rather than a manual assessment of the situation.

The solution can be configured for different numbers of production lines and plant layouts. A plant may assign more weight to delivery time or to the number of orders completed during one trip.

The customer is currently testing the solution under target operating conditions. The tests cover parameter settings, route accuracy, and system response to changing priorities.

After testing is completed, the customer will be able to assess the reduction in production downtime, the number of required trips, and line-side inventory levels.

The project was developed for an automotive manufacturer, but a similar model can also work in many other industries, including machinery manufacturing and heavy industry. It is especially relevant in plants where one warehouse supplies many assembly workstations and a delay in one delivery affects later stages of production.

A similar mechanism can also be used in aerospace manufacturing, including the assembly of complex components, the management of multiple product variants, and deliveries to selected production zones.

In both cases, the system must use actual routes, travel times, access rules, and workstation demand. Based on these data, the algorithm can organize task order, limit excess material near the line, and reduce empty trips.

We will assess how the current plant layout, order priorities, and task assignment method affect delivery times. We will help design a solution that reduces the risk of workstation shortages, limits unnecessary travel, and uses the equipment already operating in the plant.

Let’s discuss intralogistics in your manufacturing facility.