Challenges

Price Differences and Data Management Issues for Identical Specification Parts

As companies grow, the issue of purchasing identical specification parts at different prices arises.
Each vendor provides part information in various catalog formats, complicating data integration management, and manual input and differences in input methods hinder data consistency.
As a result, procurement personnel spend a lot of time comparing and analyzing part specifications and prices, leading to repeated purchases of the same parts at different prices.

AI-Based Part Similarity Analysis

Automatic Analysis and Clustering of Part Similarity

Mellerikat's part similarity analysis solution analyzes the similarity of complex columns within table data and automatically clusters items with high similarity.
It transforms unstructured part data into structured data and maximizes data usability and efficiency by analyzing similarities between parts using AI technology.
With powerful preprocessing capabilities, it structures technical specification data for purchased parts such as circuits and semiconductors, and provides customized similarity calculation methods, greatly enhancing the efficiency of the part management process.

Data Standardization and Preprocessing

Structuring Unstructured Part Data

It standardizes unstructured data of various part attributes to create a structured dataset suitable for AI analysis.
Various information such as numerical, range, and size data can be converted into unified units for standardization.
The automated preprocessing feature allows customers to add or expand new items themselves, providing high flexibility.

Easy and Intuitive Similar Part Search

Automatic Search for Parts Similar to Specific Parts

Based on the preprocessed data, when a user selects a specific part, AI automatically finds similar parts by comparing key attributes.
Without needing to know complex formulas or algorithms, users can check the list of similar parts with just one click,
making it very easy to explore alternative parts or compare prices.
The search results for similar parts are provided in clusters, allowing users to see various alternatives at a glance.

The results of the part similarity analysis are provided through the Splunk Dashboard.

Support for Design and BOM Creation

Optimal Part Recommendations with Just Specification Input

When designing a new product or creating a Bill of Materials (BOM), simply inputting the key specifications of the desired part allows AI to recommend
the most suitable parts based on the analysis results. Without complex search conditions,
users can quickly find parts that meet the required performance and conditions, greatly enhancing design efficiency and accuracy.
This feature can be utilized in various situations such as new product development, selecting alternative parts, and cost optimization.

Efficiency in AI Model Development and Operation

Simplified AI Development and Operation with ALO-ML

Mellerikat's ALO-ML dramatically reduces the complexity of developing AI models needed for part similarity analysis.
It provides a container environment by packaging Python modules, sample data, and code,
establishing a consistent development and deployment environment with Docker images.
ALO-ML automates the complex similarity calculations for large-scale part data,
allowing developers to focus on algorithm development and performance optimization.
Analysis time is shortened, accuracy is improved, and efficient data-driven decision-making is possible.
In particular, it supports rapid and efficient project progress by reducing the AI solution development period by 30%.
ALO-ML sets a new standard for AI solutions for part similarity analysis.

Automation Based on MLOps and Reflection of Latest Information

When the part catalog is updated, the model training and deployment pipeline automatically operates to
quickly reflect the latest part specifications and price information.
Utilizing AI Conductor and Edge Conductor,
the part similarity analysis process is operated systematically and efficiently,
automating the deployment, updating, and real-time monitoring of AI models to enhance operational efficiency.
This allows procurement personnel to make fast and accurate decisions based on the latest data-driven analysis results.
Additionally, it reduces monitoring work processes by 75%, and through similarity analysis between parts
and optimal price proposals, it secures price competitiveness.

Integration with Splunk

AI-Based Part Price Optimization through Splunk Integration

Through the integration of Mellerikat and Splunk, the part similarity analysis solution can be utilized more effectively.
By using Mellerikat for Splunk available in the Splunk Marketplace, users can easily integrate Splunk with Mellerikat.
Users can send Splunk data to Edge App with simple SPL (Search Processing Language) commands,
receiving real-time results of AI-based part similarity analysis and price optimization.
Additionally, users can adjust the weights of specific part attributes in real-time or instantly check analysis results based on various conditions, greatly enhancing work efficiency and decision-making speed.
Through integration with Splunk, companies can maximize the effects of data-driven strategic purchasing and cost reduction.