Robots at Estonia’s largest ready meal producer to cut labour cost by 20% and boost in-factory logistics efficiency by 10%
Kulinaaria OÜ has been preparing food since 1999. The largest product groups are salads and ready meals, complemented by confectionery and bakery products, desserts, banquette sets and sandwiches. The company’s products are mainly sold under private labels in several store and petrol station chains. Its production premises in Tallinn, Estonia, turn out more than 10 tonnes of product per day, with the 300 employees creating an annual turnover of some 20 million euro. A significant expansion of the facilities is scheduled to start operation in the first half of 2020.
Complex in-house logistics to be resolved
Transport and material handling are the key processes on Kulinaaria’s production floor, tying up a large part of the human workforce.
Because some 6,000 boxes of goods perday need to be moved between different locations within the factory, the objective of the experiment was to improve the intra-logistic process by automating the transport of boxes. Deploying AGVs (Autonomous Ground Vehicles) and using L4MS tools for the experimental study, such as OPIL (Open Platform for Innovations in Logistics), were deemed an expedient solution to optimise both the material handling process and the routes along which the material is transported.
The experiment focused on three main transport routes:
- Delivering empty boxes between the washing facilities and intermediate storage
- Transporting boxes filled with finished products to the warehouse
- Transporting raw materials to intermediate storage
The solution was expected to improve intra-logistics by automating these three delivery routes, from raw material pickup to the finished goods transported to storage. By improving indoor logistics, Kulinaaria hoped to reduce waiting time in production, improve on-time delivery and help reduce irregularities during the in-house transport process.
Communication between system components is essential
The purpose of the experiment was to test how AGVs behave in actual production conditions and whether they perform the desired actions in a satisfactory way. It was also tested how the communication between AGV, sensors and OPIL works within the company’s production facilities.
The first challenge was to construct the 3D simulation model, a virtual twin of the facility, and connect it to OPIL modules. The second challenge was to implement and test the solution on Kulinaaria premises.
The automation used several OPIL modules. The production map and the locations of the AGV bases were described in the MOD.SW.SP module that referenced the 3D simulation model of the production floor (virtual factory), developed on the Visual Components 4.1 software. AGV bases in the washing area were defined as MOD.IOT.SAN showing whether the boxes are at base through using an infrared sensor.
The whole process was to be executed with the help of OPIL SAN module integration, combined with communication between the ERP system, sensors and AGVs.
3D digital twin helps implementation
Results showed that it takes around six months to develop the whole solution, starting from the 3D simulation to testing in the production facility. During the experiment, the creation of the digital twin 3D simulation and the use of OPIL helped reduce the installation time of robots. Moreover, the proper implementation of OPIL and 3D simulation will likely help reduce the implementation time at other sites.
The cost of automation
In total, Kulinaaria’s production requires 5–6 AGVs. One AGV can serve three production lines at a utilisation rate of 76%.
The cost of one AGV system, including all supporting hardware and software, is around €43,500. The overall cost for implementing five robots works out at around €300,000.
Efficiency up, errors and labour cost down
The results of the experiment point to several benefits.
Defects such as irregularities in the quantity of boxes, and therefore goods, transported, boxes transported to a wrong location, and incorrect goods transported are most likely to be reduced by 10%.
On-time delivery of boxes to the right location within the facility is most likely to be increased by 5%.
The speed of transport measured by the number of boxes transferred per time unit will most likely increase by 10%.
The labour cost for moving the boxes will most likely decrease by 20%. The labour can be used for more productive work.
Result: Replicable solution!
Almost all goals and expectations of the experiment were met. The intra-logistics process becomes more efficient because less time and workforce are needed for transportation. Additionally, employees do not need to leave their workplace as raw materials and finished goods are moved automatically.
Mobile robots provide more flexibility and a better possibility to phase in investments according to increases in production capacity.
Kulinaaria’s judgment is that their solution can be replicated in other companies that work with similar production and business processes.
Tallinn University of Technology, TALTECH
Project leader and technology provider
Experiment Testing Facility