Assessing post-conflict electric power supply reliability in low voltage distribution networks of Aksum Ethiopia

Abstract

Conflicts in different regions of the world have had significant consequences on electric power supply appliances. This study was conducted in Ethiopia’s Aksum town low-voltage (LV) electric supply lines using reliability indices, including SAIDI, SAIFI, CAIDI, and EENS, before, during, and after the post-period. Accordingly, the findings indicate that severe reliability problems occurred due to the conflict, which resulted in power interruptions (both forced and operational), infrastructure damage, the displacement of skilled engineers, and inadequate recovery efforts. The study introduced a mixed auditing mechanism, analyzed historical power outage data, and evaluated performance-based reliability indices. Results indicate that before the conflict relatively stable power supply, with manageable outage durations and energy losses. However, during the two-year conflict (2020–2022), the average SAIDI value peaked at 1,626.4 h, the CAIDI flowed to 2,507.31 h, and EENS intensified to 10,480.52 MWh, reflecting significant disruptions. Six months post-conflict, significant recovery efforts improved these metrics, though supply stability challenges persist. The resulting reliability of the post-conflict reliability indices exceeds international benchmarks, with SAIDI over 1–5 h, SAIFI above 0.5, and CAIDI over 5 h. Despite post-conflict improvements, high SAIFI values show supply instability. Strengthen infrastructure, improve maintenance, and enhance auditing for reliable power.

Supplementary Information

The online version contains supplementary material available at 10.1038/s41598-026-35599-y.

Introduction

The conflict in Tigray, which started in late 2020¹, has severely impacted the region’s infrastructure, including the electrical power supply system. The electrical grid suffered extensive damage, including high-, medium-, and low-voltage components.

This study focuses on the post-conflict impact on power reliability in Axum town’s low-voltage (LV) feeder lines. Key components of the grid were damaged, leading to poor power reliability and quality. Severe damage affected protection devices, insulators, poles, towers, transformers, transmission lines, and substations. Power reliability auditing helps to identify system vulnerabilities, enabling preventive maintenance to reduce downtime, extend equipment lifespan, and lower costs. The Ethiopian Electric Utility (EEU) can use this approach to assess equipment performance and reliability.

Our audit found that power reliability problems in Axum’s LV feeder lines during and post-conflict were mainly caused by utility-side disturbances due to faults (L-G-F, L-L-F, L-L-G-F), destruction of protection devices (neutral wire, earth line, circuit breakers, and transformer fuse boxes), and poor maintenance. These issues have caused frequent power interruptions, appliance damage, and poor customer satisfaction.

This paper has presented a technical audit of power reliability based on the physical damage of feeder line equipment, lack of maintenance, which led to permanent and temporary interruptions in Akum’s LV feeder lines during the conflict and post-conflict. We identify key reliability problems, outline damaged feeder line components, and compare results against standard benchmarks to validate the findings.

Impact of the conflict on the electric grid system

Studying low-voltage (LV) feeder reliability is crucial for stability, maintenance, and operational efficiency. Sustainable equipment protection and high-quality power supply are essential in today’s technology-driven environment. The application of regular audits and maintenance ensures smooth and reliable electrical operations. Electric utilities and consumers focus on improving power quality and reliability cost-effectively. A power supply reliability assessment evaluates system stability, availability, and performance by analyzing downtime, outages, and indices like SAIFI, SAIDI, and CAIDI.

The fault analysis determines effective isolation of short circuits, loose line connections, and overloads, while power quality parameters such as voltage dips & swells, harmonics, and transients are reviewed for optimal system performance analysis.

We analyzed power supply availability in Ethiopia’s Tigray region, focusing on Mekelle, Shire, and Aksum before and during the conflict¹. As a result of this study, the continuous power blackout from 2020 to 2022 has caused an economic crisis in the entire community of the Tigray region¹. While power supply increased in Addis Ababa and other cities in Ethiopia, Tigray’s capital city, Mekelle, declined during the entire period of the conflict¹; even sometimes, poor-quality power supply availability led to the worst grid plug-in equipment. Since the start of the conflict in November 2020, grid-connected customers in the Tigray region have suffered a significant power supply outage. Since the end of the conflict in November 2022, power supply in Tigray was in a recovery process, but has not yet returned to the pre-conflict level by June 2023¹.

During the phases of the conflict, Aksum town’s power grid faces reliability problems. To address this, we have computed Power Reliability Auditing (PRA) for systematic power supply reliability assessment and maintenance. PRA evaluates power consistency, availability, and resilience using SAIFI, SAIDI, and CAIDI indices, and the load side power supply unavailability, EENS. By identifying weaknesses in generation, transmission, and distribution, PRA enables targeted improvements, enhancing overall power reliability. Digital processing and systematic analysis play key roles in ensuring a stable power supply.

We conducted this study in the historical place of Aksum town, northern Ethiopia, about 974 km by road and 588.7 km by air from Addis Ababa, located at 14.740°N, 38.730°E, and 2,124 m above sea level², in Aksum town with a population of over 120,000.

Electric equipment damage

The conflict caused severe destruction of critical transmission and distribution infrastructure in Aksum town and nearby towns. According to the EEU Shire district office report in 2022, over 1,000 poles and 43 transformers, including protection devices, were destroyed and rendered as indicated in Table 1.

Table 1.

Damaged infrastructure during the conflict.

Equipment	Amount	Location
Transformer	43	Enticho-Shire
	86	Total in Tigray (23 out of use)
Poles	1000	Entisho-Shire only
Transmission lines	33 kV (Replaced)	Shire-Sheraro (123 km), Shire-Endabaguna (20 km), Nebelet-Adwa-Aksum-Shire (168 km), Enticho-Gerhoranay (km),
	66 kV (maintained)	Shire-Aksum-Adwa (83 km)
	15 kV (replaced)	Edagaarbi-Adwa (42 km)
	230 kV (Maintained)	AbyiAdi-Werei (28 km)
Substation	Tekeze (maintained)	Generation (Hydropower plant)
	Mekelle (maintained)	Distribution
Protection devices, insulator devices, distribution boards, e.t.c.	In the entire Tigray	A lot in their number

This is unusable, as indicated in Fig. 1, leading to extensive and frequent power outages that disrupt the electricity supply and reliability.

Fig. 1.

Cement pole line support failure during the conflict.

The indicated outages are exacerbated by the inability to carry out timely repairs and maintenance. As indicated in Figs. 1 and 2, the conflict in Tigray has destroyed line protection devices, cement poles, wooden poles, and transmission lines.

Fig. 2.

The feeder line support wooden poles and bush insulators failed due to the conflict.

As shown in Fig. 3, feeder line transformers, support lines, and lighting arresters were also damaged during the conflict, and due to a lack of preventive maintenance and loose connections in the line, most transformer dropout fuse protection devices have burned and been destroyed.

Fig. 3.

Feeder line transformer support and drop-out fuse damage.

The images collected during the conflict, as shown in Figs. 1 and 2, and 3, illustrate severe damage to the transmission and distribution infrastructure. Figure 3 depicts the failure of a cement pole line support, likely due to direct attacks by weapons, prolonged exposure to harsh conditions, or lack of maintenance during the conflict time. The collapse or severe damage to these poles resulted in disrupted power transmission, leading to frequent outages in affected areas. As indicated in Fig. 2, the wooden poles appear damaged or knocked down, while the bush insulators, essential for electrical insulation and stability, and are broken or burned. The destruction of transformer drop-out fuses indicates compromised overcurrent protection, further weakening the distribution system. Collectively, these images illustrate the devastating impact of the conflict on the region’s electrical infrastructure, leading to prolonged and widespread power outages in Aksum and nearby towns.

Power reliability problems in the study area

Power reliability problems at the low voltage distribution feeder lines are mainly caused due damage to electric equipment during conflict period faults, loose connection of neutral line, improper groundings, interruptions, and load variations^3–5. Power interruptions that lead to poor power reliability are caused by equipment failure, natural disasters, and grid instability. In Ethiopia, equipment such as transformers, circuit breakers, and switches malfunction due to aging or lack of maintenance, leading to widespread outages^6–12. Natural events like storms, floods, and lightning strikes regularly damage power lines and substations, causing prolonged interruptions. Transmission line faults, often triggered by environmental factors like fallen trees or ice, further disrupt the power supply^6–12. Additionally, grid instability, which occurs when unbalance supply demand and capacity or there is an imbalance between generation and consumption, results in blackouts or rolling outages.

Furthermore, deliberate vandalism or sabotage of critical infrastructure in the selected area should severely affect the electricity supply. Addressing these problems leads to enhancing grid stability, modernizing equipment, improving infrastructure, investing in backup power solutions, and ensuring a rapid response to restore reliable power.

A power interruption can be temporary, sustained, or momentary. Mainly, sources of power interruption include physical damage to power lines, load shedding, temporary self-clearing faults in the power system, lightning strikes, and utility switching operations. Basically, reliability has to do with the complete loss of voltage, current, and voltage sine waves, total electric interruptions, not just variations of the power waves. Typically, reliability does not cover sags, swells, impulses, or harmonics¹². Reliability indices typically consider such aspects as connected load, number of customers, frequency of interruptions, and amount of power (kVA) interrupted.

Common fault types in Aksum LV feeder lines

In low-voltage electric distribution systems, fault sources that lead to interruptions in service and potential damage to equipment. Understanding the different types of faults is crucial for effective management and mitigation of power reliability strategies. The common types of faults that lead to power interruption (forced and operational) and poor reliability are described as follows^13–17,36.

Distribution permanent short circuit (DPSC): happens when conductors make direct contact, creating a low-resistance path for current. Causes include insulation breakdown, equipment failure, or external factors like tree branches. The fault persists until cleared by protective devices and can lead to severe voltage dips, transformer damage, and cascading failures. Mitigation involves protective relays for fast fault detection and isolation, as well as vegetation management to reduce tree-related outages.

Distribution permanent Earth fault (DPEF): occurs when a phase conductor remains continuously connected to the ground due to insulation failure, cable damage, or environmental factors like moisture. This fault persists until manually cleared, causing voltage drops and triggering protective devices like earth fault relays. It poses safety hazards to the public and maintenance personnel. Mitigation includes regular insulation testing, maintenance, and ground fault protection devices to isolate faults quickly.

Distribution temporary short circuit (DTSC): Results from brief conductor contact due to transient events such as switching operations, wind-induced sway, or temporary debris. It self-clears quickly but can cause momentary voltage fluctuations. While it typically does not require repair, frequent occurrences impact system reliability. Mitigation involves advanced protective devices, fuses, and circuit breakers designed to handle transient faults efficiently.

Forced interruptions: typically caused by faults, operational interruptions, which are planned outages, deliberate power blackouts during conflicts, maintenance, or upgrades. The causes of these interruptions range from equipment malfunctions, severe weather, conflict, and accidents for forced interruptions, to necessary maintenance for operational disruption. The consequences can significantly affect customer satisfaction and service reliability, as well as lead to financial implications for utilities due to penalties or loss of customer trust.

Distribution temporary Earth fault (DTEF): is a transient connection between a phase conductor and ground, often due to momentary insulation breakdowns from environmental factors like lightning or moisture. These faults usually clear themselves, but can cause operational issues if recurrent. To mitigate risks, regular line inspections, vegetation clearance, and outage detection systems are essential for rapid response.

Electric power supply reliability auditing

Power supply reliability auditing is a systematic process used to assess the quality of electricity delivery, efficiency, and consistency within a given system. It involves evaluating and identifying vulnerabilities, key reliability indices, and recommending improvements to minimize power disruptions and enhance service quality.

In Axum city, due to the absence of regular inspection and maintenance, electric grid power components such as power meters, distribution transformer dropout fuses, and loose connection neutral line are very common, as shown in Fig. 4.

Fig. 4.

Common power reliability problems overview in Aksum town: (a) power meter; (b) load directly connected to the distribution transformer.

We conducted the main reliability indices for this research study^18,19, which include:

1. SAIDI (System Average Interruption Duration Index) – measures the total outage duration per customer per year.

2. SAIFI (System Average Interruption Frequency Index) – indicates the average number of power interruptions per customer annually.

3. CAIDI (Customer Average Interruption Duration Index) – represents the average time required to restore service per outage event.

4. ASAI (Average Service Availability Index) – evaluates the percentage of time electricity is available in a year.

A power reliability audit is a structured evaluation process used to assess the consistency and dependability of the electricity supply in a given system. It helps identify weaknesses, improve infrastructure resilience, and enhance power quality.

The most common types of power reliability auditing mechanisms include preventive audit, corrective audit, compliance audit, performance-based audit, and financial audit^20–22. But here in this research study, we have used the corrective and performance-based auditing methods.

Review of literature

A review of literature on power supply reliability in our world, specifically in developing and underdeveloped Sub-Saharan countries, reveals many challenges and advancements in the sector. Here are some key points and findings:

S. Mulugeta²³: Much of Ethiopia’s power infrastructure is outdated and prone to failures. This includes transmission lines, substations, and transformers, which are often insufficient to meet the growing electricity demand.

B. Abdisa²⁴: The frequent and prolonged power outages result in significant economic losses, particularly for industries that rely on a continuous power supply for operations. Manufacturing sectors, agriculture, and service industries are particularly impacted by unreliable electricity.

Morgan D. Bazilian²⁵: Studies how traditional power-system planning fails in fragile and conflict-affected states and argues for risk-aware, resilience-focused planning (favoring distributed, modular generation and staged investments). His work supports incorporating conflict uncertainty into planning/auditing frameworks.

Neha Patankar²⁶ developed a stochastic electricity-planning model that explicitly includes the probability of conflict damage (South Sudan case), showing that hedging against conflict risk often favors distributed/solar solutions over large centralized investments. Very relevant methodologically for including conflict risk in LV feeder audit scenarios.

Hisham Zerriffi²⁷: Early, influential work arguing that distributed generation/cogeneration can reduce vulnerability of power systems in conflict environments; his papers show distributed systems’ theoretical advantages during instability.

Javier Salmerón²⁸: Seminal researcher on adversarial/interdiction modelling of power grids; developed bilevel optimization approaches to identify grid components whose loss would cause maximal disruption (useful for vulnerability ranking of critical LV/MV nodes under targeted attacks).

Jacob Mays²⁹: Analyzes how market structures and incomplete risk-trading affect resilience to rare extreme events; his work links market design to social resilience and can inform audit items about contractual/financial vulnerabilities during crises.

U. Khalid³⁰: Recent empirical work showing how internal conflict worsens energy poverty; valuable for connecting outage/reliability metrics to household-level impacts in post-conflict settings.

A. Al-Ain (Al-Akori)³¹: Mixed-methods studies documenting how conflict damages both water and power networks and how utilities and communities cope; highlights institutional and community dimensions auditors must include.

Mohamed Almaktar³²: Empirical and feasibility work on restoring operational reliability in Libya via local solar and hybrid microgrids after conflict; practical case studies and simulation approaches you can cite for technical/pathway recommendations.

This research article focused on the identification of power reliability issues in LV feeder lines using auditing mechanisms for Aksum city during the conflict time in Tigray.

Description of the grid system

The electric power supply voltage level in Ethiopia includes 550 kV, 400 kV, 230 kV, and 132 kV primary transmission, with 66 kV and 45 kV sub-transmission and 33 kV and 15 kV distribution systems³³. The primary transmission carries high-voltage power over long distances from the generation plants, while sub-transmission and distribution systems step down voltage to distribute to the end users or customers. The Aksum town substation has two power transformers serving industrial, residential, and commercial users. It receives a 132 kV feed from Adwa substation, and a 230 kV feed from Tekeze hydropower plant, but only the 132 kV transformer is operational, as the 230 kV transformer is non-functional³³.

Aksum distribution substation is supplied by Tekeze hydropower plant and Adwa substation via a 230/132 kV transmission line. It operates with a 40/50MVA transformer, stepping down to 15 kV for distribution through two feeders and two reserve switchgear panels, further reduced to 380/220V for customers. The system follows a radial topology with a single bus bar scheme, which is cost-effective and simple but lacks operational flexibility. In case of a bus fault or circuit breaker failure, the entire system must be de-energized. Despite its limitations, this layout is easy to operate and requires minimal protection. Aksum substation has two transformers, but only the 132 kV (40/50MVA, 132/15 kV) unit is functional, serving industrial, residential, and commercial customers. The 230 kV transformer is non-operational. The system follows a radial feeder topology, with an impedance voltage of 10.33% and operating temperatures of 60 °C (oil) and 65 °C (winding)³⁴.

Methodology

In this study, to accomplish the required result and make conclusions, we made a preliminary study area survey on the electric grid network to assess the condition of low-voltage (LV) distribution networks affected by the Tigray conflict and to evaluate the accessibility and reliability of available data sources before formal data collection. Next to this, we collected necessary primary and secondary data, and consulted different books and internet searches. To understand and obtain the technical reliability performance analysis for the specified site, it is important to conduct data collection and analysis, identify findings, and compare with standard benchmarks. To achieve this, we conducted an assessment study on the existing feeder lines of the selected site. Accordingly, we determined the main power reliability key performance indicators, including SAIFI, SAIDI, CAIDI, and EENS (Expected Energy not Served). These indicators are selected based on the data availability in the SCADA system of the feeder line substation (Aksum substation). The main activities done throughout the study are included in Fig. 5.

Fig. 5.

Overall research methodological framework of the study.

As we collected data from the main substation, the number of transformers, their total MVA capacity, and the feeder line length from Aksum town to the end are summarized in Fig. 6. Feeder K_06 has a high-power transformer capacity, while feeder K_04 covers a larger transmission line length and more distribution transformers.

Fig. 6.

Aksum city feeder line transformers, capacity, and feeder length.

Almost all the entire city loads (commercial loads, industrial loads, and residential loads) are supplied from the feeder line with K_06 as indicated in Fig. 6. The feeder line with K_04 is supplying the nearby towns with a distance coverage of 45 km, including Mahbere deigo, Daero hafash, Semema, Edaga berhe, Tsatsilo, Teregay fancha, Chila, Chemo, and Wikro. The load profile of these feeder lines is summarized in Fig. 7.

Fig. 7.

Load profiles of K_04 and K_06 feeder lines (residential, commercial, and industrial).

Aksum substation is supplying above 14,890 customers, including 12,222 residential, 2,545 commercial, and 123 industrial connections. Feeders K_04 and K_06 supply are serving around 97% (14,443.3 customers) of the total load, making them critical to the system as indicated in Fig. 7. We selected these feeders for evaluating yearly interruptions, forced and total outages, and various fault types.

Feeder lines power interruption profile

According to the data collected from the Aksum distribution substation, the main causes of momentary (unplanned) power interruptions include Distribution Temporary Earth Fault (DTEF), Distribution Temporary Short Circuit (DTSC), Distribution Permanent Earth Fault (DPEF), Distribution Permanent Short Circuit (DPSC), and General Fault/ blackout (GF). In addition to this, the planned (forced) power interruptions, specifically during the conflict in Aksum, as the main cause of power interruption.

We collected detailed power interruptions before, during, and post-conflict to make a detailed study on the impact of the conflict in Aksum town on electric power supply reliability, as indicated in Table 2.

Table 2.

Total power interruption profiles of Aksum substation feeder lines.

	Before the conflict (three months)		During the conflict (2020–2022)		Post conflict (Six months)
Feeder line	Freq.t Interruptions, F/day	Interruption durations, D (Hr)/day	Freq.t Interruptions, F/day	Interruption durations, D (Hr)/day	Freq.t Interruptions, F/day	Interruption durations, D (Hr)/day
K-04	2.59	1.62	1.36	7.56	3.37	1.65
K-06	1.49	0.93	0.62	6.40	1.89	1.50

As indicated in Table 2 before the conflict, Aksum’s grid, despite rising demand, aging infrastructure, had moderate interruption frequencies. Feeder line K-06 experienced 1.49 interruptions and 0.93 h of downtime, while K-04 feeder had 2.59 interruptions per day with 1.62 h of downtime daily. These interruptions indicated the grid’s weaknesses, as outdated infrastructure, including transformers, power lines, and switchgear, struggled with increased demand. Inadequate maintenance and limited fault detection contributed to frequent, though brief, outages, highlighting the grid’s vulnerability.

During the conflict (2020–2022) indicated in Table 2, the frequency of interruptions dropped significantly to 0.62 interruptions per day on K-06 and 1.36 interruptions per day on K-04, but the duration of outages increased drastically. The K-06 had 6.40 h of outages daily, and the K-04 feeder saw 7.56 h of downtime per day. While the frequency of interruptions decreased, it is due to direct physical damage, infrastructure neglect, or grid blackout. Since the grid was not fully functional during the conflict, fewer minor faults were recorded, but when faults did occur, they led to prolonged downtimes, as repairs were difficult to carry out in the conflict zone. The lack of simple maintenance and the severe damage to critical infrastructure meant that each fault required more time to repair, thus increasing the duration of interruptions.

In the post-conflict period shown in Table 2, interruption frequency increased, with K-06 experiencing 1.89 interruptions per day and K-04 experiencing 3.37 interruptions per day. The number of frequent interruptions is lower during the conflict than before, showing that grid recovery was ongoing. Outage durations decreased significantly, with K-06 having 1.50 h of downtime daily and K-04 having 1.65 h. The rise in interruptions reflects the fragile recovery of the grid, as much of the infrastructure was still being rehabilitated. Some grid areas remained unstable, causing more frequent interruptions, while electricity demand grew as the city recovered. Additionally, restoring the grid faced challenges, including integrating old and new components, leading to occasional faults and longer repair times. Despite shorter outages, full restoration to pre-conflict conditions was not achieved.

The higher frequency of interruptions before and after the conflict is due to the aging grid, equipment failure, and variable demand. During the conflict, fewer but longer outages occurred due to the grid’s limited operation and repair challenges. Post-conflict, interruption frequency slightly increased as the grid recovered, while durations decreased, indicating progress in rehabilitation but ongoing restoration challenges.

Data analysis and discussion

The term reliability in the utility context refers to the amount of time end users are totally without power for an extended period of time (i.e., a sustained interruption). The three basic load point reliability indices determinants usually used are the average failure rate, the average outage time, and the average annual unavailability or average annual outage time.

We depicted the main power system reliability indices SAIDI (System Average Interruption Duration Index), SAIFI (System Average Interruption Frequency Index), and EENS (Expected Energy Not Supplied), considering failure rates found at the feeder lines under study, using the upcoming equations³⁵.

The index SAIDI measures the total duration of power outages for customers over a specific period, expressed in minutes or hours. We depicted the SAIDI using:

1

Where: r_i is the outage time for each interruption event, and N_i is the number of interrupted customers for each interruption event during the reporting period at load point i. N_T is the total number of customers served for the area.

SAIFI measures the frequency of outages experienced by customers over a specific period, expressed as the average number of interruptions per customer. We depicted it using:

2

Where: λ_i is the failure rate at load point i, and N_i is the number of interrupted customers for each interruption event during the reporting period at load point i. N_T is the total number of customers served for the area.

CAIDI: We assumed that the interruptions affected all customers (as no specific information about customers interrupted is given) to calculate using:

3

EENS is a measure of the total energy not supplied, but the term is often used in the context of future reliability and expected outages. We calculated this using:

4

Where: La(i) is the average load given by La(i) = Ed/t; and Ed is the total energy demanded in the period of interest t for fault interruption i.

Using (1) to (4), we found the value of the main reliability indices as indicated in Table 3.

Table 3.

Summary of reliability indices across conflict phases for both feeder lines.

Customer-oriented reliability indices	3_months pre-conflict	2020 (9 months during the conflict)	2021(during the conflict)	2022 (9 months during the conflict)	6 months post-conflict
SAIDI (Hour/Customer/year)	134.16	1455.51	557.42	2866.29	162.14
SAIFI (Interruption/Customer)	0.6045	0.5632	0.8697	0.6671	0.8502
CAIDI (Hour/Interruption)	221.9	2584.357	640.93	4296.64	190.708
EENS (MWh) for both feeders	39.52	7595.98	208.0	23,637.57	47.78

We audited the power supply reliability in Aksum feeder lines significantly pre, during, and post conflict conditions. Before the conflict, nearly stable operations with manageable power supply outages, while during the conflict severe disruptions of power due to infrastructure damage and maintenance challenges. After the conflict, the recovery efforts focused on restoring power reliability through repairs, appliances, and system upgrades, but failed due to an unreliable power supply.

The reliability indices indicated in Table 3 show significant variations across different phases of the conflict, reflecting the impact of conflict and subsequent recovery on the electricity distribution network. These indices include SAIFI, SAIDI, CAIDI, and EENS, which collectively provide insights into system reliability and customer service levels. We calculate each value considering failure rates, outage time, and interrupted customers, intended for the three-phase scenario are summarized in Table 3.

Result analysis and discussion

As we calculated the impact of the conflict on power supply reliability challenges indicated in Table 3, the SAIDI values reveal significant outage duration fluctuations. Three months before the conflict, the SAIDI value was 134.16 h, indicating comparatively few interruptions. During the 9-month conflict time in 2020, the SAIDI value increased to 1,455.51 h as indicated in Fig. 8, which was due to the severe impacts. In 2021, the SAIDI value dropped to 557.42 h, which shows partial improvements as shown in Fig. 8. Yet, in 2022, its value increased to 2,866.29 h, reflecting severe reliability deterioration due to infrastructure failures. Six months post-conflict, SAIDI improved to 162.14 h, indicating recovery efforts, but demanding further improvements; this value is greater than before the conflict.

Fig. 8.

Reliability indices profile during the phases of the conflict.

The SAIFI, which measures outage frequency per customer, also fluctuated as indicated. Three months before the conflict, SAIFI was valued at 0.6045, indicating moderate reliability. During the 9-month conflict time of 2020, it slightly decreased to 0.5632, suggesting stable, frequent interruptions despite prolonged outages. In 2021, SAIFI rose to 0.8697, which indicates more frequent disruptions. Minor recovery practices were made in 2022 as SAIFI declined to 0.6671. However, six months post-conflict, SAIFI rose again to 0.8502, indicating persistent reliability problems still in the existing feeder line.

From the CAIDI index value indicated in Fig. 8, three months before the conflict, CAIDI stood at 221.9 h, indicating relatively long but manageable outages. During the 9-month conflict time of 2020, CAIDI rose to 2,584.36 h, indicating very long restoration times due to maintenance challenges. In 2021, it was improved to 640.93 h, signifying better response efforts. However, in 2022, the CAIDI value fell to 4,296.64 h, marking the worst performance, which is due to conflict-related damages. During the six months post-conflict, CAIDI dropped to 190.71 h, reflecting improved operational efficiency, but still not reliable.

From the load point index EENS shown in Fig. 8, the three months before the conflict EENS were 39.52 MWh, indicating a little stability. During the 9 months of the conflict time of 2020, it increased to 7,595.98 MWh, reflecting severe interruptions (forced or operational) on the power supply. In 2021, its value diminished to 208.0 MWh, showing improved outage management. However, in 2022, the EENS value rose to 23,637.57 MWh, showing severe supply challenges. Six months post-conflict, EENS fell to 47.78 MWh, showing significant recovery and stable energy supply, but greater than before the conflict.

Result validation analysis against standard benchmarks

Since, no fixed global benchmark (depends on system size) for the EENS, but values above 100 MWh/year for small urban feeders indicate serious reliability risk, we selected the reliability indices SAIFI, SAIDI, and CAIDI for comparison analysis against the standard benchmarks as indicated in Table 4.

Table 4.

Comparison of the reliability indices against standard benchmarks.

Region/Country Type	SAIDI (hours/customer/year)	SAIFI (interruptions/customer/year)	CAIDI (hour/interruption)	Source
Developed countries (e.g., USA, Germany, UK, Canada)	0.5–2 h (best), up to 5 h typical	0.2–1.0	1–2 h	IEEE Std. 1366–2012; EPRI; ENTSO-E reports
Emerging economies (e.g., India, Malaysia, Brazil)	10–100 h	0.5–3.0	2–5 h	CIGRÉ, World Bank, Asian Dev. Bank
Sub-Saharan Africa (SSA, developing)	100–200 h	2–10	10–200 h	World Bank (AICD); AfDB reports
Ethiopia (normal peacetime average)	100–200 h	1–3	100–200 h	EEU internal & MoWE performance reports
Aksum City (this study)
Pre-conflict	134 h	0.6045	221 h	Aksum substation SCADA system
Peak-conflict	2,866 h	0.7 (average)	4,297 h
Post-conflict	162 h	0.8502	191 h

We considered the standard values of the above Table 4 for the comparison analysis as indicated in Fig. 9, and accordingly, the SAIDI and CAIDI values during the peak-conflict time are very high. The SAIFI remained roughly stable (0.6– 0.85 interruptions per customer), meaning the number of outages didn’t change much, but each outage lasted much longer. This indicates long-duration blackouts rather than frequent short interruptions.

Fig. 9.

Comparison of the reliability indices with standard benchmarks.

We compared the calculated reliability indices indicated in Table 3 with international standards set in Table 4, including the World Bank (Africa Infrastructure Country Diagnostic (AICD)) and African Development Bank (AfDB). Reports for Sub-Saharan Africa (SSA, developing) and results indicate, Ethiopia’s Aksum town has faced significant reliability problems. SAIDI values in highly reliable networks typically range from 1 to 5 h per customer annually, with up to 10–100 h in the Emerging economies countries (e.g., India, Malaysia, Brazil)^36–38. The extreme SAIDI values observed, particularly in post-conflict (162.14 h), reflect extreme missing from the standard values. Similarly, the SAIFI benchmarks for reliable networks range from 0.1 to 0.5 frequent interruptions per customer per year^17,36–38, whereas Aksum’s values (0.5632 to 0.8697 interruptions) suggest a higher-than-standard outage frequency. The CAIDI benchmarks generally target 1 to 2 h, with up to 2–5 h considered high in extreme conditions^17,36–38. Aksum’s post-conflict value is extraordinarily high, indicating serious inefficiencies. From the EENS, it is aimed to minimize energy losses, yet the determined 47.78 MWh value in post-conflict, which is a greater value than before the conflict that signals major reliability failures of the feeder line.

Conclusions

The conflict severely damaged Aksum’s electrical grid, destroying transmission towers, transformers, feeder lines, and protection devices, leading to extensive outages. Weapon fire and poor maintenance further weakened wooden poles, insulators, lightning arrestors, and line structures, which led to reduced grid power supply reliability even in the post-conflict conditions. Damaged distribution transformer, dropout fuse protection devices, and poor preventive maintenance caused operational and forced faults.

We made reliability audits on the main causes of poor power supply reliability challenges during the conflict and made comparisons and analysis with post-conflict and before the conflict conditions. The worst period was 2022, with extensive conflict-related damage. However, post-conflict improvements in restoration and maintenance have led to some extent better reliability, though Aksum’s electricity supply still falls limitation of international standards.

To enhance reliability, efforts made focus on key indices values such as SAIDI, SAIFI, CAIDI, and EENS. Strengthening infrastructure, conducting systematic audits, and adopting strategic operational improvements are crucial. These steps will ensure a stable power supply, improve customer satisfaction, and minimize economic losses from outages, ultimately helping Aksum town to achieve global electricity reliability benchmarks.

In this study we faced limitations such as full data unavailability during the conflict, no detail electric damage assessment data recorded results in the entire region and difficulty to get damage related photos during the conflict.

Supplementary Information

Below is the link to the electronic supplementary material.

Abbreviations

CAIDI

Customer Average Interruption Duration Index

DPEF

Distribution Permanent Earth Fault

DPSC

Distribution Permanent Short Circuit

DTSC

Distribution Temporary Short Circuit

DTEF

Distribution Temporary Earth Fault

EC

Ethiopian Calendar

EEP

Ethiopian Electric Power

EEU

Ethiopian Electric Utility

EENS

Expected Energy Not Supplied

L-G-F

Line to Ground Fault

L-L-F

Line to Line Fault

L-L-G-F

Double Line to Ground Fault

LV

Low Voltage

MVA

Mega Volt-Ampere

PRA

Power Reliability Auditing

SAIDI

System Average Interruption Duration Index

SAIFI

System Average Interruption Frequency Index

AICD

World Bank Africa Infrastructure Country Diagnostic

Author contributions

(A) Hagos Gebrekidan Berhe: He is the principal investigator for this research article. And identify the problems, indicate objectives, review related works, develop a method, compute results, analyze, and discuss. (B) Dr.Milkias Birhanu Tuka (PhD): He has contributed as an Editor and reviewer in the work. (C) Gebrehewot Miruts Kebedew: He has contributed to this research work as a co-investigator. Participated in the activities of the work.

Funding

Funded by Aksum University, Aksum, Ethiopia.

Data availability

All data generated or analysed during this study are included in this published article [and its supplementary information files].

Declarations

Compliance with ethical standards

This research article is focused on the reliability challenges resulting from the conflict zone, and it is ethically acceptable for similar works.

Competing interests

The authors declare no competing interests.