Frequently Asked Questions (FAQ)

Information and answers about Peerox, MADDOX, and our services.

1. About MADDOX

In many manufacturing companies, rapid problem-solving during production downtime depends on the knowledge of individual employees. MADDOX makes this knowledge available in a structured way and combines it with current production data so that causes can be identified more quickly. Calls to maintenance, time-consuming searches in folders and databases, or lengthy dialogues with chatbots are thus a thing of the past.

MADDOX was developed to make valuable experiential knowledge permanently available and to put people in production back at the center. The goal is to make knowledge from the minds of individual experts usable for the entire company, to enable machine operators to resolve minor malfunctions, and to relieve the burden on maintenance.

MADDOX is suitable for manufacturing companies with complex systems, automated production, and recurring malfunctions. Companies with a high proportion of experiential knowledge particularly benefit from structured knowledge preservation.

No. In traditional knowledge databases, the user searches for the appropriate content. In contrast, MADDOX analyzes current data and automatically suggests relevant content.

2. Functionality and AI

MADDOX analyzes machine and process data as well as historical events. On this basis, malfunction patterns are identified and appropriate causes as well as corrective actions are suggested.

The software compares data patterns from current situations with the patterns of past events to which knowledge entries have been assigned. . This allows probable causes to be displayed in prioritized order.

Yes. Based on previous experience, documented problem solutions, and current process data, MADDOX suggests concrete measures for troubleshooting. Employees can then provide feedback on whether the suggested solution was successful. This requires only one click.

Yes. Every documented cause and solution expands the knowledge base and improves future suggestions. This improves the quality of suggestions and MADDOX continuously learns along with the company.

Then a new knowledge entry (“knowledge card”) can be created. The new knowledge is then available for future malfunctions.

Yes. MADDOX can identify relationships within production lines and identify the actual cause of a cascading failure.

No. The focus is on analyzing and resolving existing malfunctions. However, MADDOX can be combined with predictive maintenance modules from other providers to provide knowledge for avoiding predicted malfunctions.

The AI analyzes data patterns and links them with experiential knowledge. This automatically suggests appropriate causes and solutions.

3. Benefits and Value

MADDOX shortens fault analysis times, reduces downtime, and makes experiential knowledge permanently available. Additionally, experts are relieved and onboarding times are shortened.

The most important benefits include:

  • Faster root cause analysis and resolution of machine malfunctions
    • Faster resolution reduces Mean Time To Repair (MTTR)
    • Root cause resolution increases Mean Time Between Failures (MTBF)
  • Reduction of system downtime
    • Less scrap
    • Higher system availability
  • Preservation of experiential knowledge
  • Relief for experts and maintenance
  • Relief for experts and maintenance
  • Traceable documentation of events (actual causes) and measures

The exact savings potential depends on industry, production volume, and system structure. As a guideline, approximately half of the difference between experts and inexperienced employees can be saved. Even a few avoided downtime hours per year can generate significant economic benefits. Many companies particularly benefit from shortened fault analysis times and higher system availability.

No. MADDOX does not replace employees. The system supports machine operators in decision-making and troubleshooting. MADDOX secures existing knowledge from skilled workers and makes it available more quickly.

Yes. New employees often cannot yet draw on experiential knowledge. With MADDOX, they gain quick access to documented knowledge and can solve typical problems independently. Especially during skilled labor shortages and generational transitions, MADDOX helps to provide experiential knowledge in a structured way and significantly shorten onboarding times.

Knowledge is documented in the form of structured knowledge cards and linked to the corresponding machines and systems. This ensures it is retained even during staff changes.

Knowledge cards and problem solutions can be used at multiple sites. This allows proven knowledge to spread more quickly throughout the company.

4. Machines, Data, and Integration

In principle, almost all machines with a controller (PLC) can be connected. Older machines can also be integrated.

If no usable data can be provided, automatic analysis and search by the algorithm is not possible. The knowledge management functions can still be used. However, the database must then be searched using keywords.

Machine status information, error messages, and relevant process data are required. The more meaningful the data, the better the analysis.

A minimum of 30 data points per machine is recommended. Ideally, 100 to 300 data points are used for detailed analysis. Even large systems, such as paper machines, with several thousand data points can be connected without problems.

No. MADDOX uses existing interfaces and does not require access to the PLC code.

In addition to PLC data, MES, SCADA, ERP, historian, and maintenance data can be integrated.

Supported protocols include OPC UA, MQTT, S7 (RFC 1006), Modbus TCP, and other industrial standards. Additionally, customer-specific integrations are possible.

Yes. It is important that there is a traceable relationship between the data and the occurring malfunctions.

Yes. Knowledge cards can also be created and used for such machines, even if automatic analysis is not possible.

Yes. Content and trained models can be transferred to comparable machines. In addition, machine groups with a shared AI model can be created. This allows MADDOX to learn from all machines simultaneously and correspondingly n times faster.

5. Implementation and Operation

We offer a structured implementation process in which system integration, training, building a knowledge database with approximately 40 entries, and training the algorithm are completed within eight weeks. The prerequisite is that all necessary conditions are met, such as a suitable and well-documented machine data interface.

Yes. Many companies start with a single machine or line to realistically assess the benefits first.

Our pilot project is a complete package with rental hardware, software licenses, and the required training, so that no additional effort and costs arise. Approximately 80 to 100 work hours are required on the customer side, distributed among different people in various roles.

This depends on the frequency of specific malfunctions. Recurring patterns can be reliably identified after several repetitions, i.e., approximately 4-5 assignments.

After technical modifications, accuracy may initially decrease. The system adapts through new data and feedback.

Acceptance arises primarily through practical benefits in daily work. Peerox additionally supports change management and implementation.

MADDOX is continuously developed. Updates are provided approximately 4 to 6 times per year in coordination with the customer.

Peerox supports implementation, integration, and training.

6. Use in Daily Work

MADDOX can be used on PCs, laptops, tablets, smartphones, and touch panels. The application runs in the browser and can be used flexibly.

Yes. If an HMI has a web interface, MADDOX can be integrated there.

A connection to the MADDOX server is required. However, no internet connection is needed for an on-premises installation.

Knowledge cards can be displayed in different languages because MADDOX has an automatic translation function.

MADDOX does not have its own voice control. However, voice functions of the end devices can be used.

Yes. Content can be searched and filtered by various criteria.

Yes. Images and videos help to clearly document solution approaches and work steps.

Fault identification based on image analysis and data from camera systems is currently being investigated in pilot projects. Even today, images can be used as part of knowledge cards.

7. Knowledge Management

The content comes from experienced operators, technicians, and maintenance technicians. This means the knowledge base is built on real practical experience.

All registered users can create new knowledge cards. Approval is granted by experienced personnel who have specific rights in the software.

Editors review new content before approval. This ensures the quality of the knowledge base.

In principle, knowledge content is available to all users. This supports knowledge transfer.

Yes. Users can comment on content and share experiences.

8. Data Protection and Security

The data can be operated on-premises or in suitable cloud environments. The operating model is determined together with the customer.

The data belongs to the respective creator or operator. Ownership rights remain with the customer.

Authorization concepts, encryption, and security mechanisms protect sensitive information. Implementation is carried out according to customer requirements.

No. Customer data is not used to train third-party AI models.

The focus is on production-related information. Inferences about the individual performance of individual employees are not intended.

Companies can define access rights individually. Roles and permissions can be managed in a targeted manner.

MADDOX supports established authentication and authorization concepts. In addition, company policies can be taken into account.

9. IT Systems and Enterprise Integration

MADDOX can be operated on-premises or in cloud environments. The selection is based on security and infrastructure requirements.

Yes. The solution supports integration into existing IT and OT landscapes.

Yes. SAP systems can be connected to efficiently send information from MADDOX to SAP.

MADDOX has its own user management and can be connected to existing identity management systems.

Yes. Existing authentication solutions can be integrated.

Yes. Data can be exported and integrated into other systems via APIs.

10. Cost-Effectiveness

Yes. Companies can estimate the potential savings based on their own KPIs.

Customers report reduced downtime and fewer ad hoc interventions. The actual benefit depends on the specific use case.

11. Industry Use

Yes. MADDOX has been assessed for use in the pharmaceutical environment and classified as not GMP-relevant. The SAP interface of MADDOX is GMP-validated. The solution supports traceability and structured documentation and is therefore suitable for regulated environments.

Yes. Industry-specific requirements and validation processes can be taken into account.

12. About Peerox

Peerox develops MADDOX, an assistance system for industrial production. The focus is on combining production data with experiential knowledge. The company is based in Dresden.

Interested companies can arrange a demonstration or start with a pilot project. This allows the benefits to be evaluated based on their own use cases.

This site is registered on wpml.org as a development site. Switch to a production site key to remove this banner.