Over the last five years, Indonesia’s economic growth has exceeded 6% annually due to business expansion and industrialization. This has had a major impact on the electricity sector, which has been one of the essential pillars of the country’s economic expansion. Electricity demand has grown more than 8% annually since 2010, which has resulted in network expansion and increased use of distribution network assets.
Perusahaan Listrik Negara (PLN), a state-owned utility, holds the monopoly for the distribution of electricity to approximately 52 million customers. PLN has had to focus on expanding its distribution network while increasing the network capability, taking into consideration asset management to minimize the risk to reliability and establish lifetime maintenance standards. In response to these challenges, PLN has embarked on the implementation of distribution enterprise asset management (DREAM) by initiating a pilot project in PLN Bali. This asset management discipline aims to optimize the three conflicting drivers — namely, performance, cost and risk — throughout the life cycle of a distribution asset.
Plan of Action
PLN commenced research into asset management practices by considering combining the objectives of time-based maintenance (TBM) and condition-based maintenance (CBM) to achieve an efficient method of asset maintenance. New concepts —including asset class, health index and frequency multiplier — were introduced to CBM, and TBM included the statistical analysis on the mean time to failure of a distribution asset.
Ensuring a strong availability of assets in the long term is an important part of plant maintenance. Planned maintenance is performed to avoid system outages or the breakdown of existing system assets. In addition to asset repair or replacement cost, planned maintenance often results in much higher costs due to claims from commercial and industrial customers that experience a loss of production.
One CBM methodology design to ensure strong availability of assets is the health index methodology, which focuses on the inspection activity and subsequent corrective action. At PLN, the CBM methodology has been implemented using the SAP model plant maintenance (SAP PM). PLN has hired Accenture Indonesia to assist in the implementation of its enterprise asset management SAP PM project.
Health Index Methodology
The high-level key business processes of health index methodology are as follows:
- An asset inspection work order is created automatically before the scheduled inspection date.
- The inspection process can be in the form of a hard-copy work order from SAP or a soft file (Excel) of the work-order inspection downloaded from SAP. This soft file can be completed manually using a mobile application.
- The inspection result is processed by the utility in SAP as a measurement document using manual input in SAP into an Excel file. This is only used for inspection work orders and not corrective-action work orders.
- Based on the measurement document, the health index engine calculates and determines whether corrective action is required.
- If corrective action is not required, the health index engine will generate the next inspection schedule for processing.
Currently, PLN has only applied this asset management procedure to two distinct assets: distribution transformers and medium-voltage overhead lines.
Screening and Inspection
The two main phases in the methodology are asset screening and inspection activity. Asset screening of distribution assets is obtained based on the historical database, which uses records of permanent and temporary disruption or breakdown within the last two years, for a certain period. It also considers the asset usage level, which is the average load on the distribution feeder over a predetermined period. The risk level, which varies per asset classification, is also considered.
Inspection activity is performed if the health index for the asset is classified as active and waiting for inspection. Work- order inspection is created automatically within a certain interval before the scheduled inspection date to assist the maintenance planner in preparing for the inspection process. PLN’s headquarters can specify the interval between the work-order inspection creation date and the inspection date to meet business requirements. Currently, the work order is created automatically 10 business days before the scheduled inspection date. Work orders can be requested manually by maintenance staff, if deemed necessary.
Health Index Engine
Based on the measurement documents, the health index engine processes the actual condition of each asset twice daily by considering the results of the inspection. Results are then routed to maintenance staff. Depending on the results of the frequency multiplier in the health index calculation, the asset then can be subject to either next inspection date (if no remedial work is required) or corrective action required, which then generates work orders that could include non-replacement or replacement of the asset.
Bali Pilot Project
The DREAM system was initiated in PLN’s Bali region in early 2014 and centered on three different locations, namely, North Bali, East Bali and South Bali. Specifically, the scope of the pilot project included phases 1 and 2 of the inspection activity, which was then followed up by maintenance action in accordance with the asset health index. The complete procedures have been performed on two distribution assets, 11 20-kV overhead line feeders and 80 20/0.4-kV distribution transformers. The SAP system runs the work-order management system for the inspection activity and follow-up actions.
The pilot project resulted in a major reinforcement program for PLN to repair and replace the medium-voltage overhead lines and medium-voltage/low-voltage distribution transformers on the existing distribution network in Bali. Following completion of the DREAM pilot project and investment in the repair and replacement of distribution assets, there has been a continuous improvement in reliability and performance of the overhead lines and distribution transformers.
Because of the reduction in the number of overhead line faults, PLN has increased energy sales by reducing the energy not sold. The improved performance of the assets also has had a significant positive effect on the system average interruption frequency index (SAIFI) and system average interruption duration index (SAIDI) of the complete distribution network of PLN Bali to some extent.
PLN Indonesia implemented the DREAM system with the objective to increase the efficiency and effectiveness of asset maintenance. The DREAM system identifies the conflicting drivers of performance, cost and risk to optimize maintenance activities. Several core parameters, such as asset classification and health index, play an essential role in establishing these priorities. In addition, the sequential procedures for maintenance activities, established in the DREAM system, ensure every maintenance activity on an asset is carried out accordingly, based on the real condition of the asset.
Finally, the DREAM system has greatly improved PLN’s management of many PLN assets to ensure their performance can be improved in the most efficient way. The Bali project is under evaluation before the utility embarks on a rollout to other branches in accordance with PLN’s project road map, currently being defined.