Monitor process data from the voice application in real time: With Lydia Warehouse Intelligence (LWI), topsystem, an EPG company, has a smart analytics tool for the Lydia Voice software in its portfolio that allows users to continuously monitor the performance of the system in real time. This includes, for example, the status data of the voice hardware used, such as the battery life of the mobile voice computers, but also environmental data, such as the WLAN field strength or the response times of the system to the server.

LWI is one of the first smart data analytics solutions for voice processes on the market. The evaluations are based on previously defined KPIs, which are stored and continuously monitored in the BI tool. The tool gives users of LydiaVoice a real-time overview of the status of the warehouse in relation to the voice software and hardware used as well as the infrastructure. This is possible because the voice devices send the status data of the hardware - such as battery life, but also environmental data such as WLAN field strength or system response times - to the server.

LWI collects and stores these values over a defined period of time. The tool then structures and analyzes the data and makes it clearly available in a dashboard for further evaluations. In the example of batteries, LWI continuously monitors their runtime, charging cycles and maximum performance. On this basis, conclusions can be drawn about the remaining service life and older batteries can be replaced in a targeted manner.

The tool also collects data for the continuous evaluation of speech recognition quality. The individual recognition score of voice users indicates how well employees are understood by the voice system. In the latest version, LWI has an extended speech recognition analysis (word-phrase statistics): Based on the words spoken, the tool generates statistics so that it is easy to identify whether there is potential for optimization with regard to the recognition of individual words or entire phrases. The use of LWI eliminates the need to analyze log files from various sources, which was previously time-consuming. LWI continuously processes all data and makes it available for analysis purposes.

Basis for predictive maintenance
Warehouse managers who control internal processes and therefore need to be informed about the current status of the most important warehouse KPIs at all times benefit from LWI, according to the manufacturer. Bottlenecks, weak points and optimization potential in the voice installation can be identified at an early stage and localized more easily. In this context, LWI also forms the basis for predictive maintenance: the permanent monitoring of the analyzed data material means that any need for action can be identified before it becomes critical. For example, the capacity of the batteries can be constantly monitored and they can be replaced in good time.

AI components in the next development stage
The targeted evaluation of all data is essential for the continuously high performance of the entire voice installation. In future, LWI will be supplemented by AI components that enable differentiated monitoring of response times between the voice system and the higher-level system (WMS, ERP). This enables LWI to learn how long certain processes, such as server queries, usually take. Response times can vary on a customer-specific basis, depending on the type of interface and infrastructure. Machine learning enables LWI to recognize patterns and adapt possible escalation levels accordingly. LWI is available as a SaaS solution and can be used completely without customizing.