PLANT LOGISTICS
Just be there without attracting attention.
A washing machine has 1,800 parts. In production, each one has to be in the right place at the right time. A mammoth logistical task – ideal for the use of AI.
Hardly anyone consciously notices things that everyone uses every day as a matter of course. This is especially true for mobile phones but also for household appliances. Dishwashers, washing machines, dryers, vacuum cleaners, ovens, and refrigerators all make everyday life easier. They are only noticed when they are broken or missing. This is the experience of appliance manufacturers. But they have to deal with thousands of individual parts that make up washing machines and dishwashers. To assemble them in the right order, they must arrive at the right place in the production line at the right time. If a part is missing, the worst that can happen is that production grinds to a halt. This is where the importance of a functioning internal logistics system becomes clear. Artificial intelligence-based software systems help companies keep the flow of materials flowing smoothly. In this way, they avoid unwanted conspicuities caused by missing materials.
Plant logistics consists of two processes: external material deliveries and internal material transports. Large manufacturers have to handle more than 100 trucks per day in the area of external material deliveries. In addition, depending on the size of the plant, there are several hundred internal transports to move material from the warehouse to production.
It is when a part is missing from the assembly line that the importance of in-plant logistics becomes apparent.
A question of sequence: external delivery logistics.
The challenge in delivery logistics is quickly unloading the arriving trucks at the plant. This means having the right people with the right equipment in the right place at the right time. In addition, truck traffic in front of and inside the plant must be organized to avoid traffic jams. Also, the process must be efficient, meaning employees, means of transport, and loading points must be used evenly. Knowing exactly when which truck will deliver which material is critical to smooth operations. To do this, manufacturers allocate time slots and take into account all the conditions that affect speed: When is the freight needed? How many trucks need to be unloaded at which docks? How many trucks can move through the facility at the same time? And how many people and what resources are available and when?
Only software based on AI and mathematical optimization can keep track of all these conditions simultaneously and calculate the optimal allocation of delivery dates based on the overall situation
You have more AI in your home than you think!
Did you know that AI controls many household appliances? A washing machine measures the amount of laundry and how dirty it is, then optimizes the amount of water and spin speed. As a result, the machine uses less water and electricity. AI also supports the user’s daily life with personalized program suggestions, automated cleaning processes, and easy-to-understand error messages.
Real-time optimization is critical.
As soon as a truck arrives at the plant, this type of software also calculates the optimal throughput sequence of all trucks currently in the plant. In addition to the scheduled delivery date, the system knows the current status of all resources at all times. Based on this information, the truck drivers receive precise instructions on which loading points to visit and in what order. Real-time optimization is crucial. No matter how well you plan ahead, there is always the chance that unforeseen events, such as traffic jams or breakdowns, will cause trucks to arrive late. In this case, the AI calculates a new dispatch plan for the changed situation within a few seconds. All parameters are taken into account, such as parking and ramp utilization, the current status of loading operations, or possible load priorities. A human being alone could never accomplish this task at the required speed. However, the human still has the last word. Dispatchers can manually intervene in the planning process at any time and override the system’s decisions at their discretion.
Every day, there are millions of options, but only one ensures a smooth flow of materials while considering all constraints. Every day, schedulers search for the famous needle in a haystack - and find it, thanks to AI!
Internal transport organization: a question of correct allocation.
Internal transport organization is no less complex. It is characterized by delivering the right material to the right place at the right time and in the right quantity in response to a request from production. Each time a new material is requested, a decision has to be made as to which driver will carry out a transport order with which vehicle. Punctuality is just one criterion. In addition, all vehicles should be used evenly and cover the shortest possible distances while avoiding empty runs. For the allocation of transport orders, this means that it is not always the next best truck that receives an order but rather the one with which a subsequent order can best be connected.
Good plant logistics are like a good washing machine: they’re there, they do their job, and they don’t draw attention to themselves.
With more than 100 transport orders a day, these conditions alone make scheduling so complex that a human would be overwhelmed. This is why companies rely on software with intelligent optimization algorithms. These are often self-learning algorithms. This means that the software can take into account transport orders even before they are placed. Suppose the system knows, for example, that the transport of a certain stock material always takes place at 4 p.m., and that this order could be completed in the morning, thus avoiding an empty run. In that case, the algorithm makes this decision independently in the background.
VIDEO: AI IN EVERYDAY LIFE
Plant logistics
In this video our colleague Jennifer Stead, AI Catalyst at INFORM, explains how AI is helping appliance manufacturers organize their factory logistics.
AI finds the needle in the haystack.
If you take a mathematical look at the factory logistics of a major appliance manufacturer. In that case, there are an almost infinite number of sequences in which you could handle all the tasks that occur each day. Even with ten transport orders, the number of possible sequences can exceed 3 million. The trick is to find the one that ensures smooth material flow and efficient processing, taking into account all the constraints. In this mammoth logistical task, schedulers search for the proverbial needle-in-a-haystack every day – and find it, thanks to AI. When it comes to internal logistics, appliance manufacturers are like consumers with a washing machine. It’s there, it’s working, and it’s not attracting attention.