Solutions: Business Solutions
Solutions: Business Solutions
The amount of data generated in companies during the value chain and beyond is increasing immeasurably. The data can be real-time data as well as historical data and can come from a variety of sources. Examples include enterprise data such as ERP and CRM, machine data, log files, text, voice & video data, process data or location data.
However, all this data only becomes usable when information is extracted from it. To do so, it is necessary to intelligently merge the data and visually process it in real time. The knowledge gained from this is very valuable and contributes to the improvement and automation of processes or the creation of new business models. Your very decisive competitive factor!
CANCOM Analytics uses the latest technologies and a wide range of analysis methods to address precisely this issue, enabling you to implement almost any conceivable application scenario: from the visualization of actual conditions, to forecasting models for process optimization, to autonomous, self-learning systems. The following diagram provides an overview of the different levels of CANCOM Analytics.
At the beginning of an analysis process, there are always data sets from raw data. However, what can be derived from these and what possibilities arise from them depends on the analysis methods. We distinguish between the following analysis stages:
Descriptive analytics creates transparency about current data and processes.
Diagnostic analysis finds causes for events and developments.
With the help of historical data and special analysis methods, future events and conditions can be predicted.
Prescriptive analysis provides you with data-supported recommendations for action for your company-relevant decisions.
With the help of analytical methods, intelligent automation of subtasks and processes is made possible.
Real-time analysis of your acquired data provides you with valuable insights and knowledge. The prerequisite for this is to define in advance exactly what data is collected and for what purpose.
Once the relevant data has been analyzed, you can make informed decisions based on it in order to achieve specific goals. One possible goal, for example, is to optimize processes to save resources, improve the company’s performance and achieve higher profits.
Depending on the decision, the corresponding action finally takes effect. This is to realize the intended goals, such as process optimization.
To properly exploit the potential of data and generate added value, data is analyzed in six phases. CANCOM uses the standard procedure model “Cross Industry Standard Process for Data Mining”, CRISP-DM for short.
1. Business understanding
Create understanding on the business background of the question.
2. Data understanding
Create understanding of the data.
3. Data preparation
Pre- and post-processing of data.
Modeling using the different methods.
Evaluation and review of results.
Provision and application of the results in ongoing operations.
The analytics model used depends on the purpose of the application. In the context of data analysis, different models are used for modeling and algorithm selection depending on the use case and analysis level. These range from simple threshold analyses, cluster analyses and machine learning to neural networks and deep learning. CANCOM Analytics guides you to your perfect analytics scenario.
Industry-specific application scenarios
Analytics solutions can be applied to all industries and processes. In the following, two examples from the production and retail sectors will show you what the application scenarios can look like in concrete cases.
Predictive maintenance enables you to anticipate and avoid failures of machines and production lines. Accordingly, you can adjust your maintenance schedules and carry out maintenance operations more flexibly. With Predictive Maintenance you improve your production processes, the quality of your products and strengthen the competitiveness of your company. In addition, you increase the satisfaction of your customers and save costs.
Retail Analytics can be used to evaluate customer movements within or in the vicinity of your store. For example, the analysis provides insights into how long and where exactly the customer spends time during his visit. In this way, you gain valuable insights into the effectiveness of promotions and window displays. These results can then be used for insightful comparisons between individual locations or different measures.