{"id":9077,"date":"2024-11-11T10:15:28","date_gmt":"2024-11-11T10:15:28","guid":{"rendered":"https:\/\/www.gsma.com\/get-involved\/gsma-foundry\/?post_type=gsma_theme_resources&p=9077"},"modified":"2025-09-19T12:23:34","modified_gmt":"2025-09-19T12:23:34","slug":"nokia-predictive-hardware-analytics-services","status":"publish","type":"gsma_theme_resources","link":"https:\/\/www.gsma.com\/get-involved\/gsma-foundry\/gsma_resources\/nokia-predictive-hardware-analytics-services\/","title":{"rendered":"Nokia: Predictive hardware analytics services"},"content":{"rendered":"\n

Challenge:<\/strong> Mobile network operators face unplanned downtime risks as hardware faults disrupt critical services, relying on reactive processes for resolution. Central operations centre reactions often involve costly field visits and repairs, with around 25% resulting in unwanted field visits (considering 100k+ sites which generates 20k+ hardware related field visits and repairs annually). Traditional monitoring methods struggle to identify and resolve issues, leading to service disruptions, especially for applications demanding continuous connectivity, leading to unplanned outages and urgent interventions. This results in higher costs, increased emergency truck rolls, and adverse impacts on ESG commitments.<\/p>\n\n\n\n

Solution:<\/strong> Proactive process to identify and resolve hardware faults: Nokia Predictive Hardware Analytics (PHWA) is a telco industry first, patented<\/a>, service developed in collaboration with Nokia Bell Labs that uses data analytics and AI to forecast future health and performance of Mobile Radio Network (RAN) equipment in three dimensions; being site, module and card level. The solution collects and processes configuration, performance, fault management, and RAN data, as well as historical records and repair centre information from Nokia CRM and ERP global sources to predict potential hardware issues before they arise and enable proactive incident resolution.<\/p>\n\n\n\n

Impact:<\/strong><\/p>\n\n\n\n

Lower Operational Costs<\/strong>: Targeted interventions through prediction saves money compared to reactive repairs. Lowers associated hardware maintenance cost by 40% which translates to 0.5 to 5 million Euros yearly savings based on network size (small\/large).<\/p>\n\n\n\n