PLN smart meters: installed units are not the measure of AMI success, complete MDMS data is
Solution case studies are scenarios based on common industry problem patterns. They do not refer to any client and are not MarkasDev project portfolio items. Our client portfolio is published separately.
Industry context
AMI is run by PLN and its partners; MarkasDev was not involved. This page relies on ANTARA and Kompas coverage and other open documents. Advanced Metering Infrastructure (AMI) and smart meters are operated by PT PLN (Persero) with partners and subsidiaries named in public sources. This page is not our portfolio. We read reporting that quotes PLN officials (ANTARA, Kompas), an ITPLN write-up of a July 2025 public lecture by PLN’s Distribution Director, and a World Bank project document discussing AMI, to unpack how a national utility replaces conventional meters with two-way metering infrastructure.
According to ITPLN (22 July 2025), quoting PLN Distribution Director Arsyadany Ghana Akmalaputri, AMI is framed as a foundation for distribution digitalisation: not merely “automatic meter reading”, but transparency, efficiency, and room for customers as prosumers. The system is described as five main components: Smart Meter (real-time consumption recording), a communications network (GSM, RF, PLC, or fibre), a Head-End System (collecting and processing meter data), MDMS (storage, analytics, integration), and a Streaming Platform that feeds operational systems such as SCADA and ADMS.
In the early 2023 wave, ANTARA quoted PLN President Director Darmawan Prasodjo that the AMI pilot, with State Grid Corporation of China (SGCC), had reached 93.54 percent of a target of about 1.2 million meters across eight provinces by October 2023 — with most units at 100 percent completion except Bali at 86.88 percent at that time. Kompas (December 2023) quoted PLN Corporate Communications EVP Gregorius Adi Trianto that the 2023 replacement covered 1.2 million customers in eight distribution parent units, was free of charge for customers, and followed clustering and the service life of existing kWh meters — not an individual request queue.
This case matters for utility leaders, infrastructure SOEs, distribution operators, and digital customer-service teams weighing smart meters, MDM, and customer channels. The hard problem is not bolting a “digital box” on a wall; it is aligning meter procurement, communications, Head-End/MDMS, field-force process, and the in-app experience — then keeping them healthy as scale grows from the first millions toward tens of millions.
Problem
Before AMI, the common operating model in a large utility is door-to-door meter reading, limited near-real-time load data, and friction when customers want to understand a running bill. PLN’s Darmawan, quoted by ANTARA and Kompas (2023), stressed that AMI enables remote real-time reads, removes the need for staff to visit only to read a meter, and opens remote disconnect and reconnect. Customers are expected to see load profiles and running bills via PLN Mobile.
The utility business problem runs deeper than “tired meter readers”. Without reliable interval data, anomaly detection (including signals of non-technical loss), faster outage recovery, and systematic postpaid–prepaid flexibility are hard to run at scale. Gregorius (Kompas, December 2023) highlighted service flexibility, faster recovery because the system can detect issues in real time, and customer privacy because staff need not enter properties as often merely to read meters.
On the customer side, meter migration raises trust issues. Kompas (13 December 2023) reported some customers feeling bills rose after the swap. Gregorius replied that AMI measurement does not change the tariff per kWh; spikes usually relate to higher usage, a failing old meter, or other factors — and customers can request a usage history at a PLN office. That marks a communication and data-quality risk familiar in AMI programmes worldwide: a more accurate new meter can “feel expensive” if the old meter under-registered.
On scale, ITPLN (July 2025) describes a ~15-year roadmap toward 34 million smart meters, with an Early Commercial phase in 2023–2024 said to have installed 1.36 million meters, then Stabilisation targeting 7.54 million (2025–2027), Ramp-Up to 16.52 million, and a PACE phase to 34 million. World Bank document P180992 cites a similar trajectory with slightly different mid-phase numbers (7.7 million by 2026; 16.7 million by 2029; 34 million by 2037), a PPP for the first 1.5 million meters, and subsidiary ICON+ managing the Head-End and MDMS. Differing figures across sources are a signal that the AMI roadmap is a living governance object — not a single finished slide.
Constraints
- AMI meter replacement in 2023 is described as free for customers and not based on an individual request queue, but on clustering and the service life of existing kWh meters — EVP Gregorius Adi Trianto via Kompas (15–16 December 2023)
- The pilot/early rollout covers eight regions/parent units: North Sumatra, Banten, Greater Jakarta, West Java, Central Java, East Java, Bali, and South Sulawesi (ANTARA also refers to South/Southeast/West Sulawesi) — ANTARA October 2023 & Kompas December 2023
- Early-wave target about 1.2 million meters; progress by October 2023 quoted at 93.54 percent, with Bali at 476,422 units (86.88 percent) still below 100 percent then — ANTARA
- ITPLN/Arsyadany narrative (July 2025): Early Commercial 2023–2024 installed 1.36 million; about 11,000 data concentrators and about 11,000 km of fibre in the early phase across eight UIDs
- Tariff per kWh is stated not to change merely because of the AMI swap; bill-spike complaints must be examined via usage history — Gregorius via Kompas (13 December 2023)
- Public architecture: Smart Meter to communications (GSM/RF/PLC/fibre) to Head-End to MDMS to streaming to SCADA/ADMS; PLN Mobile as the customer channel for load profile/running bill — ITPLN/Arsyadany & President Director quotes via ANTARA/Kompas
- World Bank (P180992): PPP for ~1.5 million first meters (availability payments); ICON+ said to manage Head-End and MDMS; mid-phase figures differ slightly from the ITPLN narrative — we note the discrepancy, we do not adjudicate it
- PLN HTML press pages we attempted on 1 October 2026 could not be read in full (SPA/404). Figures on this page come only from sources we successfully opened
Approach
Diagnose
This section is analysis: how a Diagnose phase would show up on a case like PLN’s AMI. It is not a claim that MarkasDev delivered work for PLN, SGCC, or ICON+.
Paid Diagnose is useful before a utility or SOE locks budget to “buy a million meters” without an inventory of decision rights. The questions are concrete. Who owns interval meter data — the distribution unit, the subsidiary that runs Head-End/MDMS, or the PPP vendor? Which territories enter the first wave (peak load, high losses, communications density), and what clustering criteria are transparent to field teams? What SLA covers remote reads, remote connect/disconnect, and fallback when the communications medium fails?
The pattern visible in public sources is a separation that must be mapped first: (1) devices and field installation, (2) communications network and concentrators/fibre, (3) Head-End + MDMS as systems of record, (4) integration into operations (SCADA/ADMS) and the customer channel (PLN Mobile), (5) change communication so “more accurate” is not heard as “tariff went up”. Mixing all five into one SOW without acceptance criteria often yields meters on walls that are not yet “alive” in the MDM or the app.
For teams copying the pattern elsewhere, Diagnose also tests people–process–technology readiness: SOPs for visits only on anomalies (as Gregorius described), contact-centre capacity at cluster go-live, and privacy policy for household load data. Without that, a political go-live merely moves the queue from “the reader never came” to “why did my bill rise”.
Deliver
This is also analysis. Deliver breaks the work into milestones with testable acceptance criteria.
A sequence that fits the public description: first, prepare Head-End/MDMS and communications paths before installation volume peaks (the World Bank names ICON+ on Head-End/MDMS; ITPLN names five AMI components). Second, install by UID/cluster with end-to-end tests: meter to concentrator/network to Head-End to MDMS to PLN Mobile display. Third, test remote read and remote connect/disconnect on an agreed sample. Fourth, test edge cases: failed communications medium, anomalous meters, and procedures for staff visits only for maintenance. Fifth, customer communication scripts: free of charge, per-kWh tariff unchanged by the swap, how to read the load profile in the app.
Example acceptance criteria (our analysis, not official SLAs whose second-level numbers we quoted from public docs): percentage of meters successfully registered in MDMS within X days of install; percentage of complete daily interval data; latency of load-profile display in PLN Mobile; communications incident runbooks; and reconciliation of bill complaints against before/after usage history. See the Deliver and system integration & APIs frames for similar interface-contract patterns.
What to avoid: chasing physical installation counts while MDMS or the customer channel is not ready. In the public narrative, AMI’s value appears when data flows and customers see a load profile — not when a new meter box is merely visible on the wall.
Operate
This is analysis too. After the early wave, the roadmap toward further millions of meters (Stabilisation, Ramp-Up, PACE per ITPLN/Arsyadany) turns AMI from a project into an operating factory. Operate answers who watches daily communications success rates, who handles “bill feels higher” tickets, who maintains concentrators/fibre, and which reports distribution leadership receives each week.
Published scale clarifies why Operate is not optional. Early Commercial is cited at 1.36 million meters (ITPLN, July 2025) with thousands of concentrators and thousands of kilometres of fibre; ANTARA recorded a jump of more than one million within months in 2023. At that volume, one firmware regression or one bad RF zone becomes thousands of tickets. PLN Mobile as a self-serve customer channel also needs a product owner after AMI features ship — separate from merely “download the app”.
In the Operate frame, monitoring MDMS data quality, Head-End security patches, field-force capacity for anomalies, and proactive communication when bill displays change become routine work. Integration into SCADA/ADMS (named by ITPLN) adds an operations surface: meter data is not only for billing, but for network reliability. The distribution transformation Arsyadany describes — toward prosumers and intermittent renewables — only matters if AMI data stays healthy after the go-live ceremony.
High-level architecture
This diagram is our simplification of the five components described by Arsyadany (ITPLN, July 2025) plus the PLN Mobile channel from President Director quotes via ANTARA/Kompas — not an official architecture we audited. World Bank (P180992) states subsidiary ICON+ manages the Head-End and MDMS for AMI in Indonesia — we did not verify day-to-day operational boundaries beyond that document. SGCC is named on the ~1.2 million meter pilot (ANTARA) — we did not open technical contract detail. We did not open PLN’s internal diagrams. Component names and flows above come only from public sources we successfully read.
Typical outcomes
Every figure below comes from sources we opened. Scope and dates differ, so do not merge them into one “PLN AMI ROI” metric.
Early-wave progress 2023 (ANTARA, quoting President Director Darmawan). Target about 1.2 million AMI smart meters across eight provinces; realisation by October 2023 quoted at 93.54 percent. Breakdown: North Sumatra 41,821 (100 percent); Banten 138,733 (100 percent); Greater Jakarta 175,100 (100 percent); West Java 126,926 (100 percent); Central Java 52,071 (100 percent); East Java 81,750 (100 percent); South, Southeast and West Sulawesi 55,501 (100 percent); Bali 476,422 (86.88 percent). Darmawan said more than one million smart meters were replaced in about four months. Collaboration with SGCC is cited on that pilot.
December 2023 narrative (Kompas, quoting Gregorius). Replacement covered 1.2 million customers across eight parent units in 2023; free of charge; based on clustering and meter age; communicated benefits include load profile and running bill in PLN Mobile, postpaid/prepaid flexibility, faster outage recovery, and remote digital meter reading.
North Sumatra UID (ANTARA, 25 January 2024). PLN UID North Sumatra Communications Manager Yasmir Lukman: about 42,800 AMI customers in the territory (focus Helvetia, Medan); nationally more than one million at that time; installation free and said not to change the customer service type.
Long-term roadmap (ITPLN, 22 July 2025, quoting Distribution Director Arsyadany). National target 34 million smart meters on a ~15-year map. Early Commercial 2023–2024: 1.36 million meters installed. Stabilisation 2025–2027: target 7.54 million. Ramp-Up: up to 16.52 million. PACE phase: up to 34 million. Early phase: eight UIDs; about 1.3 million customers; about 11,000 data concentrators; about 11,000 km of fibre.
Corroboration / alternative figures (World Bank P180992). PPP for the first 1.5 million advanced meters with availability payments; ICON+ on Head-End and MDMS; plans up to 7.7 million (2026), 16.7 million (2029), and 34 million (2037). Mid-phase figures differ slightly from ITPLN — we show both.
What we did not find in primary sources we opened: PLN’s operational cost savings in rupiah attributable to AMI in Indonesia (Darmawan cited Austria/Netherlands as illustrations, not Indonesian audited results we verified); an official non-technical-loss reduction percentage attributed solely to AMI in an annual report we opened in this session; Head-End technical SLAs in seconds; or the full text of PLN’s 2023 HTML press releases whose URLs we tried but could not read.
Typical or expected outcomes, not claims about MarkasDev project results.About this analysis
This page is compiled from open sources and is not a MarkasDev client project. We read ANTARA and Kompas reporting quoting PLN officials (2023–2024), an ITPLN article (22 July 2025) on a public lecture by PLN Distribution Director Arsyadany Ghana Akmalaputri, and excerpts from World Bank document P180992 on AMI, PPP, and ICON+. Every figure on this page comes from a source we opened, with the scope we state. PLN website press URLs we attempted on 1 October 2026 could not be read in full (SPA/404), so we do not cite them for figures.
Why this case matters for your organisation
Electric utilities, infrastructure SOEs, water/gas operators, and digital customer-service teams often face the same question: how do you replace mass measurement assets without breaking customer trust and without letting data stall on field devices? PLN’s AMI pattern — cluster waves, technology partners, Head-End/MDMS, a mobile channel, a tens-of-millions roadmap — is a public example that can be dissected through Diagnose, system integration & APIs, and Operate.
Reading the case through Diagnose, Deliver, Operate
Three questions in order. Who owns the data and the first-wave territory criteria? That is Diagnose. How do you test meter to Head-End to MDMS to app with acceptance criteria before chasing installation KPIs? That is Deliver. Who keeps daily communications quality, bill tickets, and platform patches healthy as scale moves into Stabilisation/Ramp-Up? That is Operate. If you are weighing an initial mapping, see also Diagnose.
Limits of this analysis
We did not interview PLN, SGCC, or ICON+. We did not audit the Head-End/MDMS. We did not test PLN Mobile on a device. ITPLN and World Bank roadmap mid-phase figures differ slightly — we leave that explicit. Energy-saving claims for Austria/Netherlands cited by the President Director are treated as comparative illustration, not measured Indonesian results. Treat this page as discussion material for leaders, not an audit report.
Sources
Accessed on 1 October 2026 (WIB).
- Institut Teknologi PLN (ITPLN), At ITPLN Campus, Director Arsyadany on Distribution Digitalisation, 22 July 2025.
- ANTARA, PLN continues accelerating AMI smart meter deployment for optimal service (quoting President Director Darmawan Prasodjo; October 2023 progress).
- ANTARA, PLN: 42,800 customers use AMI smart meters in North Sumatra, 25 January 2024.
- Kompas, How to Replace a Conventional Meter with an AMI Smart Meter, 16 December 2023 (including page 2).
- Kompas, Eight Regions That Have Replaced Conventional Meters with AMI, 15 December 2023.
- Kompas, If the Meter Is Replaced with AMI, Does the Cost per kWh Change?, 13 December 2023.
- World Bank, project document P180992 (PDF), Advanced metering infrastructure section — document link.
- Attempts to open PLN press releases about AMI (June/November 2023) on pln.co.id — full body not readable in this session; URLs noted in the local CATATAN-SUMBER without citing figures from pages that failed to load.