Author: Oskar Skrzyniarz (BiModal Planning)
In the era of rapidly developing artificial intelligence (AI), we often overlook the specific ways it can assist us. Sometimes, in the race to adopt trends, we implement AI without considering its true implications. Meanwhile, awareness of the capabilities of AI algorithms translates into the potential to enhance work efficiency in every aspect of a business, supporting humans in their day-to-day tasks.
Let's emphasize the word “every,” as though it can be seen as a hyperbole, the possibility of using various AI algorithms arises in many departments of production (both on the production side and in administration). In this article, we invite you to a peculiar Gemba Walk, during which we present various examples of AI applications. We will also briefly discuss the impact AI can have on a company's strategic, tactical, and operational levels.
The article's takeaway:
- AI adoption supports all aspects of business across i.e. industries such as production, supply chain, and administration.
- AI technologies can be used to improve decision-making and streamline processes.
- By implementing AI in business, a company may enable advanced sales forecasting and optimize logistics and workforce management.
- AI capabilities aid recruitment, improve internal communication with chatbots (e.g., IBM), and enhance financial forecasting.
Leveraging AI in manufacturing
Currently, most industrial plants use machines to streamline production processes. Robots have taken over part of the work previously done by humans. One of the most common uses is machines for palletizing operations. However, this is not the only place where the applications of AI systems can find its use.
The applications of industrial AI solutions: the Ford example
Ford has been using AI tools for over 20 years to control the quality of car components. This is done using a real-time image detection and classification system. Today, workers can take a photo of a given part with their smartphone and, within seconds, determine whether the part is defective.
In this case, everything is based on training a neural network, providing it with a set of photos, with information about whether they show a good or defective product. As a result, Ford can even identify uneven wrinkles on their seats.
The use of AI in Siemens: industry data aggregation and analysis
However, the transformative impact of AI is not just about spectacular robots, camera setups, or sensor networks. AI is transforming the background of manufacturing processes, analyzing raw data collected during production processes. Analyzing this data helps answer questions such as: what happened, where, and why.
For years, Siemens has provided early warning systems for failures. By using AI-driven Edge devices that collect data from various points in the process, it is possible to capture correlations between parameter changes and an upcoming failure.
Now, think about all the situations we know from experience from various manufacturing use cases – one unexpected breakdown paralyzes the entire plant, production line, or even the entire supply chain. How valuable would it be to prevent such situations before they occur, based on data that often already exists, buried deep in corporate databases?
AI applications across forecasting services
Let’s now move on to the flagship example of the applications of artificial intelligence and machine learning – forecasting. Our associations immediately lead to sales forecasting. However, in many cases, simple statistical models are used for this purpose (e.g., ARiMA or the Holt-Winters model).
In simple cases, forecasting based on historical data, such models will be ideal. However, the reality is more complex, and the number of units sold in July next year will depend on many factors. This situation will require so-called multivariate forecasting. Simple statistics may be insufficient in the face of dozens or even hundreds of variables.
Here, both machine learning algorithms and neural networks come to the rescue. By training them on a multivariate dataset, we ensure the models will successfully identify which of the variables we provided influence our sales.
Sales Predictions and AI Business Use Cases: Rossmann’s Network
One of the great artificial intelligence use cases of using multiple variables to predict sales is the analysis described in this article: link. The authors compare sales data from Rossmann with additional factors, such as the distance to the nearest competitor’s store or the presence of holidays. Including such variables allowed for more effective training of the neural network.
Now, think about how many data points, which we often intuitively consider important, we could now realistically use in our forecasts thanks to custom AI solutions.
AI implementation in forecasting: other examples
Let’s not forget that prediction does not end with forecasting future sales. Here are some examples:
- UPS uses a global AI solution to predict the exact delivery date of parcels, considering weather conditions, traffic, and delay history in a given region.
- Mode Logistik GmbH & Co. KG uses machine learning to estimate the number of product returns which improves warehouse and logistics planning.
Thus, the use of artificial intelligence can support forecasting not only in sales but everywhere data meets decisions – from logistics and of customer service, to production planning.
The application of artificial intelligence in resource planning
Let’s move on to resource management. This is another area where the integration of AI solutions is becoming commonplace. In fact, most modern APS (Advanced Planning and Scheduling) systems already use machine learning algorithms to support the creation of optimal schedules.
This happens not without any reason, as production planning—and resource management as a part of it—is a complex and difficult task that requires accounting for a large number of variables, such as:
- resource availability,
- setup times,
- order priorities,
- delivery dates,
- technological constraints.
We often focus solely on ensuring that the plan can simply be executed. We forget that every inefficient schedule leads to real, often high, and sometimes hidden costs.
This is where integrating AI can truly become a game changer and an invaluable support to human intelligence. By analyzing past data and identifying patterns or dependencies that would be too subtle or complex for a human to catch in a short time, AI enables companies to make decisions with greater precision and speed.
Global AI and ML solutions in resource management: use cases
What about the other aspects? After all, production planning isn’t just about sequencing orders. Fortunately, even at this stage, artificial intelligence can be beneficial.
Take Walmart, for example, which uses machine learning algorithms to optimize delivery schedules. AI helps minimize transport time while reducing fuel costs, leading to savings and increased efficiency in logistics operations.
Meanwhile, H&M uses AI to manage the workforce. With AI, they can optimally adjust shift schedules and the number of employees to current needs. This allows for better resource utilization and improved work quality. This approach is important not only in operational terms but also in terms of employee engagement and satisfaction.
The role of AI in business administration
Until now, we have mainly focused on the operational side of the business. However, it is also worth looking at the administrative side of the organization, because AI offers wide possibilities for application in this area as well.
AI use cases in the recruitment processes
This is no longer a novelty, but it’s worth emphasizing that recruitment processes are increasingly being powered by AI. AI can analyze hundreds of CVs and candidate profiles in no time, allowing for the exclusion of candidates who are not a good fit for a given position, while helping focus on those with the greatest potential.
This approach, taken by companies such as Mastercard, Electrolux, and PepsiCo, not only accelerates recruitment but also increases its precision. They also use artificial intelligence to help track progress in the recruitment process, allowing recruiters to match candidates to requirements more precisely and make better-informed decisions.
AI adoption in administration: chatbots & customer contact automation
Another interesting example of an AI application is chatbots. They support both customer contact and internal communication within a company. One example is IBM, which applies AI to automate tasks and relieve its HR department. Thanks to chatbots, employees can quickly receive answers to frequently asked questions, e.g., vacations, salaries, or internal procedures. Only more complex queries are passed on to the HR department, which significantly reduces the workload.
Both parties benefit – the HR department has fewer routine inquiries, and employees get help faster. This is a time and resource-saving solution for both sides, as well as increased communication efficiency within the organization.
AI solutions in the administrative sector: financial analysis
We cannot forget the financial analysis of company performance - another field where the use of artificial intelligence proves beneficial. Predictive algorithms enable more accurate economic forecasts, allowing for more effective fund management and better preparation for upcoming market changes.
AI applications across industries: conclusion
In this article, we have demonstrated that artificial intelligence can be applied in every aspect of running a business. Choosing the right form of AI, selecting the right data, and building a model accurate for your industry is a completely different, more challenging art. Fortunately, that’s what teams like BiModal Solutions are here for.
Therefore, if, after reading this article, you believe that what we’ve discussed applies to your business, or if you have your vision of implementing AI in your organization, feel free to contact us. We would be happy to advise and assist in implementing innovative solutions.
We also encourage you to follow our upcoming articles, in which we will show in detail how AI empowers sales forecasting and production planning – not only in theory but in real processes.