Good morning, everybody. I am Kanai, President and CEO. Thank you for joining us today for Kokusai Electric's first IR Day. We take it as a great opportunity to explain our mid- to long-term business strategy and business goals announced at the time of our IPO. We would like to explain them in detail, taking into account current market trends and the progress of our business. Here is the agenda. Today, we have a guest speaker, Mr. Lee from Applied Materials, a shareholder and collaboration partner of our company. In part one, Mr. Lee will share his view of the industry as a whole, and then I will give an overview of Kokusai Electric. After that, with Mr. Lee, we will have a Q&A session about the industry and Kokusai as a whole, followed by a break.
After a short break, part two will be a detailed explanation of Kokusai's strategy for each device, followed by a Q&A session with the presenters, who are executive officers. The session will be three hours long, but we will make every effort to deepen the participants' understanding of our company. Without further ado, please start your presentation, Mr. Lee.
Presentation. Thank you. So first, I'd like to thank Kokusai and Kanai-san for inviting me and me to be able to present Applied Materials Company and our outlook on the industry. So thank you very much. Thank you very much. When we go to the next slide. So in today's talk, we will be having some forward-looking projections and statements. Of course, with anything that's forward-looking or future-looking, there are associated risks and uncertainties. Okay. We can go to the next slide. You know, I always like to start with where Applied Materials started. We started in 1967, founded in a very small office in Mountain View, California. We sold originally industrial gases for our industry. Where are we now? We are, you know, we had just had our earnings call, so I urge you to look at the earnings call online. We just finished it on May 16th.
So the details on Outlook and specific projections can be found there in its transcripts. In a trailing 12 months, we're about $26 billion in terms of revenue. We continue to invest in R&D. It's very, very important for us. And you can see it in our patent portfolio, which continues to grow, and the number of employees numbers in the 35,000 worldwide. Okay. We can go to the next slide. Now, a couple of key messages in my talk. One, the long-term outlook is very, very strong and very, very positive. For semiconductor revenue, we believe it's going to approach $1 trillion in this 2030 timeframe. A lot of it is driven by what we see today in terms of AI, Internet of Things, EV, and renewable energy. We believe these mega trends will be very sustainable going forward in the future. The second item will be inflections.
There are many, many major inflections that are coming from our customers, and with those inflections, it creates opportunities for us to capitalize and demonstrate our value in accelerating the customer's roadmap with our broad portfolio. Applied has one of the broadest portfolios in this space, and the ability to connect across the products is one of what we think is an enabling capability to accelerate our customer's roadmaps across these inflections. The third point is innovation. We're looking at how we can innovate on how we innovate, and we're looking at broader and deeper collaboration with our customers, engaging early so we can accelerate these roadmaps for these inflections. We're also looking at how we engage with our partners. Okay. Next slide. This next slide talks about where we are today and why we have confidence in this $1 trillion in semiconductor revenue.
If you look at it today, we're in the fourth age of compute, and it's driven by AI and the Internet of Things. Okay. If we look at what that means to WFE, we can go to the next slide. So again, I talked about these end-market inflections or these mega trends around AI, IoT, EVs, renewable energy. That drives us to the $1 trillion. From Applied Materials, we have an existing playbook. We call it the PPACt playbook. We look at customers' metrics, things like power, performance, area cost, as well as time to market. And we want to use that playbook to accelerate the customer's roadmaps across these inflections that you see on the right-hand side. There's many, many inflections. We're seeing those inflections today on gate all-around, backside power, advanced packaging for high bandwidth memory.
All these and many inflections on how we enable our customers' roadmaps are very, very key. We're using our playbook to help our customers collaborate on that. If we go to the next slide, you can see also from our customer's voice, the voice of the customer, very, very key things like scaling, how do they look at sustainability, how they increase memory capacity, things of that nature are very important to them across all devices, whether it's logic, memory, DRAM, ICAPS, or Internet of Things. If we go to the next slide, one of the ways our customers are looking at this inflection and how they scale is going from two dimensions to three dimensions. This is happening across all devices. You can see it in logic with gate all-around, CFET. You see it in the 3D DRAM and then backside power.
It's already occurred in NAND. So we continue to see how they scale from two dimensions to three dimensions across all devices. And now they also are looking at it not only at a device standpoint, but at a system standpoint, how they integrate multiple devices on one module using packaging to move that forward. So another inflection that we are looking at, how we help our customers. If you go to the next slide, it's specifically discussing logic. And as you can see, the logic increasing complexity as they go from FinFET to gate all-around to CFET. Gate all-around now is ramping. And this is something that we see numerous opportunities for us to solve in this inflection. If you go to the next slide, you can see some of what we see as the challenges that our customers are looking at, what they see as the high-value problems.
If you look at inner spacer or you look at High-K metal gate, there are many opportunities here, whether you're looking at, you know, conductor, edge, tunable, and selective SiGe removal, you know, ALD deposition. These are inflections that, with our broad portfolio, we're able to connect across the portfolio, looking for a 1 + 1 is greater than 2 solution. So that's what we consider co-optimization, the ability to connect different process tools and make a more optimized module. And we're looking at that and collaborating with our customers now, as well as collaborating with our partners to enable the roadmap. If we go to the next slide, a similar thing is happening on DRAM as they go from 2-dimensional DRAM to stack DRAM for the high bandwidth memory. And then further in the roadmap is the 3D DRAM.
If you go to the next slide, you can see the same challenges that our customers are facing. If you look at the ability to stack multiple DRAM die, it will require high aspect ratio etch. It will require TSV, advanced packaging. You can already see the advanced packaging is increasing in the customer's flows. As well as, if you look forward, you can see 3D DRAM. In 3D DRAM, you have new materials that are coming in, new ability to grow that stack or etch that stack or pattern that stack. These are things that, again, collaborating with our customers will enable this inflection. It also creates an opportunity for us to look at different materials engineering, whether it's hard mask, the ability to etch this material, pattern this material. That also opens a window for us to collaborate with customers and our partners.
If you go to the next slide, I talked about this fourth wave of compute. One is powered by AI, which is advanced logic and advanced memory. But the other thing was the Internet of Things. And in the Internet of Things, it has Internet of Things, communications. It has automotive power, as well as sensor technology. And that's what Applied Materials calls ICAPS. We refer to this entire space as the ICAPS space. And it continues to be a big portion of our logic foundry going forward. If you look at the next slide, you can see the devices that we capture under what we call the ICAPS umbrella. We're looking at mature logic node, analog devices, all the way down to MEMS and optoelectronics. This type of device or customer devices is considered under our ICAPS business.
In our ICAPS business, we believe this will continue and be a significant portion of logic foundry because we continue to, one, work with our customers to further optimize their devices, as well as releasing new products into this space. If we go to the next slide, I want to talk about the third theme that I have is innovating how we innovate. We have the EPIC platform. This is the connected platform of all our labs worldwide. Obviously, we have labs in Santa Clara, but we also have labs in New York, and we have labs in Singapore. Now we're building what we call the EPIC Center in Santa Clara, actually in Sunnyvale, California. But what we're doing here is enabling a platform where we can have earlier and broader engagements with our customers, with university partners, as well as industry partners.
Again, it's a platform to help accelerate our customers' roadmaps and the time to market. The next slide talks about our recently announced net zero playbook for 2040. It's for our ESG initiatives. On this, we've had very strong ratings from external third parties. This is another collaboration that we have with our customers, as well as our suppliers, to advance the ESG roadmap to help further the industry. If you go to the final slide in closing, you know, long-term positive forecast and outlook. You know, Applied Materials is well positioned with our broad and connected portfolio of products. We want to use those connected products to advance our customers' roadmaps and inflections. Then finally, we want to take collaboration to the next level, working both with our customers, our suppliers, and our partners. Thank you very much for your attention. Thank you, Terry-san.
Thank you very much. I will give you a general overview of our company now. First of all, I would like to explain our company's management structure. As you can see, we have built a strong management structure with a management team that has long been in the semiconductor industry. We are very proud of our management team's commitment to Kokusai Electric over the years. As of April 1st, Mr. Kamiya, who had been serving as a Senior Managing Executive Officer and CFO, became a full-time director. Mr. Ogawa, who had been serving as Senior Managing Executive Officer and CTO, assumed the position of director and executive fellow. Upon resolution of the June 27th general shareholders' meeting, Mr. Kamiya will become a director as a member of the audit committee, and Mr. Ogawa will continue to serve as a director and executive fellow.
In addition, Senior Managing Executive Officer Mr. Yanagawa and Mr. Tsukada will be newly appointed as directors. This is a corporate profile. In the last 70 years of history, we have achieved strong growth together with our customers, the world's leading device manufacturers. Our products and services are of the highest quality and have been highly evaluated by our customers for a long time. Although NAND used to account for a higher percentage of sales, as shown in the pie chart below right, DRAM and logic have recently accounted for a higher percentage due to the slowdown in the NAND market and the expansion of our DRAM and logic market share. Page four is the management policy of our group. Kokusai Group's corporate philosophy is the Kokusai Electric Way, which expresses our determination to further deepen dialogue with our stakeholders and support the future with our technologies.
To realize this corporate philosophy, we are aware of our social responsibility as a specialized manufacturer of semiconductor production equipment, and we will pursue economic value and environmental and social value from both business activities and ESG initiatives. We strive to contribute to achieve the SDGs and to realize a sustainable society and the sustainable development of our group. This page summarizes our unique business model. We are a company specializing in film deposition. In the equipment business, we account for approximately 70% of our consolidated net sales. We are developing our business globally with a focus on batch deposition equipment, including batch ALD, which has the largest share of the global market, and treatment equipment, which has the third largest share of the global market.
The service business, which accounts for about 30% of consolidated sales revenue, is for relocation and modification of equipment and sales of equipment for wafer size of 200 mm or smaller. In addition to recurring business of parts sales and maintenance services, recently, sales of equipment for SiC power devices have been growing. This section summarizes batch ALD technology and treatment technology. In recent years, as semiconductor devices have become more complex and three-dimensional, the scope of application of ALD technology, which provides excellent step coverage in deposition of films with challenging and complex structures, has expanded. Since ALD technology is a cyclical gas supply process and deposition takes time, batch ALD technology, which combines batch deposition and ALD technology, is the optimal solution.
Batch ALD equipment is not simply a combination of batch deposition and ALD technology, but rather a combination of complex technology and years of expertise that only achieves high performance and is recognized by device manufacturers around the world for its high value added. Treatment technology is a highly productive solution that improves film quality with excellent isotropy and step coverage thanks to the abundant radicals produced by our unique plasma method. This section summarizes our medium to longer-term management strategy. The first growth strategy is to expand sales of our batch ALD and treatment systems to meet the increasing complexity and 3D nature of NAND, DRAM, and logic applications, and to achieve business growth that exceeds WFE growth.
The second growth strategy is to expand the scope of our business in the film deposition area by increasing sales of batch deposition equipment for mature nodes and equipment for SiC power devices, which are experiencing remarkable growth. To achieve these goals, we will focus on continuous creation of high value added technologies and products and research and development that looks 10 years into the future while strengthening our ability to make proposals to and engage with customers. The third growth strategy is to expand business in the highly profitable service business by providing services that meet customer needs throughout the product lifecycle, including parts sales and maintenance. The details will be explained by the respective business managers in part two. This is a summary on the growth drivers by device.
Details will be shared by the respective executives in part two, but drivers for materializing growth exceeding WFE markets exist in each respective area. In addition to the common trend of devices becoming more three-dimensional and complex, industry-wide trends such as HBM, silicon interposer, SiC power devices, and the localization of the semiconductor supply chain are also expected to provide further tailwinds. Here are the updated midterm goals. Based on the assumption that WFE will expand to more than $120 billion, we have set a medium-term target that we aim to achieve in three or four years starting from the current fiscal year. Specifically, on a consolidated basis, we aim to achieve net sales of JPY 330 billion or more and an adjusted operating income margin of 30% or more.
For reference, the sales composition by business segment is expected to be approximately 75% for the equipment business and 25% for the service business. The sales composition of equipment sales by application is expected to be 50% for logic foundry and others, and 50% for memory, including DRAM and NAND. The reason why the service ratio will be lower than the current ratio is that equipment sales will recover at a faster rate in the future. Although NAND sales will grow significantly as the NAND market recovers, the NAND ratio is expected to remain at around 25% as sales of GAA and other advanced logic products and mature logic products for customers in Europe, the U.S., and China expand. We will spend about 6% levels of our sales revenue in R&D each year to materialize these goals.
Although the R&D ratio appears to be declining, the absolute amount of R&D continues to grow significantly each year because sales will grow more strongly than the R&D ratio. That is all for myself.
I am Tsukada, Executive Vice President and CSO. Thank you very much. I will begin with the second part of the presentation with an overview of the market outlook and our growth strategy. The semiconductor device market is expected to more than double in size from around $300 billion in 2010 to approximately $610 billion in 2022 and is projected to grow at a CAGR of 9.5% from 2023 to 2028. As Mr. Terry has mentioned in his presentation, it is expected to reach $1 trillion in 2030.
Behind the expansion of the semiconductor device market are the growing demand for electronic devices such as smartphones and PCs, the expansion of data centers due to the spread of 5G, AI, IoT, and DX, investment in reducing environmental impact, growing demand for individual applications such as so-called GX, and industrial support measures taken by major countries. In the current global economy, demand for electronic devices has been sluggish due to the uncertain economic environment, and semiconductor device makers, especially NAND device makers, have continued to curb investment. However, inventory adjustment of semiconductor devices is progressing, and the unit price of memory devices has begun to rise, leading us to believe that market conditions have bottomed out in the first half of 2023.
We expect a full-fledged recovery in demand for semiconductor devices from the second half of 2024 to 2025 and a return to growth trend toward 2027 due to continued and accelerated technological innovation. The semiconductor equipment market has more than tripled in 12 years from about $30 billion in 2010 to around $98 billion in 2022, and is projected to grow at a CAGR of 7.5% from 2023 to 2028. Although semiconductor device manufacturers, especially NAND manufacturers, continue to restrain investment at present, we expect the demand for semiconductor production equipment to recover as demand for semiconductor devices recovers. In the medium to long term, as semiconductor devices become more complex and three-dimensional, we believe that the need for semiconductor production equipment capable of both difficult film deposition and high productivity will increase.
The NAND market is expected to recover to a level close to that of 2021 by 2026, although the size of the market in 2023 is significantly smaller than 2021, and the average annual growth rate from 2023 to 2028 is expected to be 17%. The DRAM and logic and foundry makers markets, including those for mature nodes, are expected to continue their growth trend toward 2028 with a CAGR of 8.3% for DRAM and 6.3% for logic and foundry from 2023 to 2028. We would like to explain our market share. On the left are share data from Gartner Research. The deposition field mainly includes the tube and non-tube categories, with the tube defined as batch deposition and RTP and oxidation diffusion as a treatment. The batch deposition market is an oligopoly between Tokyo Electron and Kokusai.
Compared to 2022, our share of the batch deposition market has decreased by about 10 percentage points. This is due to the fact that the batch CVD market in the batch deposition market is strong, while the batch ALD market on which we focus is shrinking due to the impact of restrained investment in NAND. As the pie chart on the right shows, our share in the batch ALD market remains at 70%, and since POR is rather expanding for us, we expect our share in the batch deposition market to recover and increase as the NAND market recovers. The three main players in the treatment market are Kokusai, Applied Materials, and Tokyo Electron. As shown in the pie chart on the right, our share in plasma modification in the treatment is down about 10 points from 2022.
But for the same reason as batch ALD, and as the NAND market recovers, we expect the plasma modification share will recover. Before going into the details of the technology, let me explain the most important challenge facing semiconductor manufacturers today. That is, the decrease in productivity of the deposition process due to the increased complexity of the device. As the figure on the right shows, 3D structures make device structures deeper and more complex. This increases the surface area required for deposition, which in turn increases the gas travel distance with a longer deposition time, and thus the productivity issue becomes more apparent. This is a physical structure issue, and for batch equipment, this phenomenon is a tailwind. Batch systems are highly productive, capable of depositing 50-100 wafers at a time, and provide a solution to productivity issues in complex deposition structures.
Here, we compare two deposition technologies, CVD and ALD. The difference between the two is where the gas reaction occurs. ALD is a cyclic process in which the gases flow alternatively, and the reaction occurs on the surface of the wafer, enabling uniform deposition with good film thickness uniformity or step coverage. While advanced devices require excellent step coverage, making ALD an indispensable technology. The challenges of batch ALD are low throughput due to the cyclical process and increased wafer cost. Therefore, the high productivity of batch equipment is a good match for ALD, and we believe this combination is optimal. Furthermore, as 3D and complexity increases, the need to deposit films on large surface areas with high aspect ratios further increases the time required and productivity deteriorates. Batch ALD is gaining attention as a logical solution to ALD productivity problems.
In the matrix on the right side, the vertical axis shows the nature of the film, and the horizontal axis shows the aspect ratio. The upper right quadrant shows more important functional films and films with higher aspect ratios. Comparing batch and single wafer, batch ALD is used in the upper right quadrant, while single wafer is often used in areas with low aspect ratio or in so-called sacrificial films with low functionality. Therefore, batch and single wafer are not necessarily in direct competition with each other, but are separated according to film type and aspect ratio. On the other hand, as device structures become more 3D and complex, we believe that the market in the upper right quadrant will grow, which will be a tailwind for batch ALD. On this page, we summarize the drivers for our future growth along with the roadmap for each of our devices.
In the short term, sales are driven by increased demand for DRAM and logic, including mature node, followed by an investment recovery for NAND. The bottoming out of NAND has finally been confirmed, and recovery in investment starting in 2025 is highly expected. In the medium term, growth will come from sales expansion for logic GAA generation, increased demand for advanced DRAM, and new products for SiC power devices. In the long term, there are major inflection points such as the shift to logic CFETs and 3D DRAMs, and we aim to achieve a balanced portfolio and medium to long-term growth by providing products and services that meet the needs in each of these areas. The following table shows the breakdown of sales revenue for 300mm equipment. For NAND, we have gained an overwhelming share of the 3D NAND deposition process.
We expect the demand for our products to recover and expand as the market recovers and the devices become more multi-layered. For DRAM, we are acquiring new POR in the highly difficult deposition process for advanced DRAM. In addition to the increasing demand for advanced devices for both HBM and existing applications, we aim to steadily expand sales by acquiring new POR in TAM, which will further expand as devices evolve. For logic and foundry, we will work to expand sales in the GAA generations where we have one POR and to further win new PORs in the GAA second generation. In addition, we will expand batch equipment sales to mature nodes globally to broaden the scope of our business. Through these efforts, we aim to achieve a portfolio of 50% in logic and foundry and others, 25% in DRAM, and 25% in NAND in the medium term.
We will now begin to explain the detailed strategies for each device. I'm the Corporate VP overseeing the system development side. My name is Oda Kei. First, allow me to explain our NAND development strategy. In structures such as 3D NAND, the superiority of batch ALD is proven by the law of physics. The left side shows the movement of molecules in deposition in a low aspect ratio trench and in a deep trench. Since gas diffusion takes longer in a high aspect ratio, meaning narrower and deeper trench, time is the rate-limiting factor for molecular movement. Then, since the cycle time must be longer for single wafer, efficiency is lost, and only batch processing can solve its disadvantages. In the graph below right, we can see that after 64 layers in the 3D NAND process, batch throughput exceeds single wafer, and the inversion of single wafer and batch occurs.
The advantages of mini batch, TSURUGI, and evolution of batch ALD are further explained here. The most important factor in the deposition process is optimal control of gas inflow as well as outflow. The mini batch system facilitates temperature and pressure control by reducing the two volumes, thereby optimizing the gas inflow and outflow. This reduces deposition time and overhead time, enabling faster deposition iterations. This is a very important point in ALD, which requires a cyclical process. As a result, we have a very high market share in 3D NAND, which requires the most complex and high aspect ratio deposition. The schematic diagram on the left shows six major 3D NAND processes. We expect to increase our market share in the remaining three processes as the number of layers increases beyond 200 and even beyond 500.
In other words, as NAND investment recovers in the future, we will benefit greatly from our already high market share, and we believe that we can achieve even higher market share and greater growth as development progresses further to more than 200 layers. Next, I would like to move on to the strategy for DRAM. On the left is a schematic diagram showing the major DRAM applications, and on the right is time, and we are in a position to benefit greatly from this growth. As high performance is even more important for HBM, which is currently the biggest trend in DRAM, and DDR5 modules used in D1B. In the D1B generation, we gain market share in three applications and also in the treatment process, especially in the embedded word line, where adoption by our major customers is advancing.
Further POR gains are expected as we move into D1C next year. As we continue to gain market share in D1B and D1C, the increase in advanced DRAMs such as HBM is more than just WFE growth. It is expected that our market share in D1C and beyond will increase in every process, including treatment, providing even stronger tailwind. Finally, I will explain the structural changes in DRAM. Currently, the DRAM structure is shifting to vertical DRAM between 2D DRAM and 3D DRAM. While the structure around the capacitors does not change much in vertical DRAM, the main change is around the gate, where the bit lines are located below the channel. This structural change requires thicker and more complex deposition, increasing the need for our equipment.
In addition, the reality of 3D DRAM is further increasing, and processes such as lateral deposition and embedding, which have never existed before, are emerging. So we expect significant growth on a scale similar to that we have experienced with 3D NAND. In the DRAM market, the strong tailwind of HBM from the need for higher performance and the growing need for devices in which we have a large market share after D1a, D1b, and D1c are supporting the strong growth of our equipment. We believe that structural change to vertical DRAM and 3D DRAM will further expand the market for our equipment. Thank you.
I am Kanayama, Senior Vice President overseeing the technology area. Allow me now to explain our logic strategy. Many people may have the impression that our batch ALD is strong in memory, especially for NAND, and that many single wafer products are used for logic.
But in logic, as in memory, the shift in device structure from FinFET to GAA and CFET is providing a tailwind for batch ALD. In logic, as in memory, the structure of devices is shifting from FinFET to GAA and CFET. On the left is a comparison of single wafer processes using plasma and our batch process using thermal. Plasma has a shorter lifetime, which makes it difficult for radicals to reach deep into the film and maintain film uniformity. Also, when byproducts are generated during the chemical reaction process, it is difficult to remove them in sufficient time with single wafer. On the other hand, thermal has a longer lifetime, so it is possible to maintain film uniformity over a longer period of time, even for structures with high aspect ratios, and batch can take longer than single wafer to eliminate byproducts.
In other words, as device structures become more complex and the need for time-consuming deposition at greater depths increases, the physical advantages of thermal-based batch over single wafer with plasma increase. A schematic diagram of logic is shown on the right. Increasing device complexity has increased the need for lateral deposition, giving rise to new processes such as hard masks and inner spacers. Within the context of this trend, our batch ALD has been steadily gaining market share in GAA-related processes. In the first generation of GAA, we won POR for inner spacers from all major customers, and in hard masks, we won new POR from one of our major customers. In the second generation of GAA, we expect to further increase our market share in gate spacers and hard masks as well. We also expect to win POR in the new gap fill process.
As a result, we expect GAA-related sales to grow at a CAGR of more than 100% over the next four to five years. Furthermore, when it comes to CFET, the aspect ratio will further increase, and the need for batch ALD will increase. We are currently working with IMEC to develop a lead in CFETs, and we are sensing a growing interest in batch ALD from the industry. In addition, as I explained earlier in the DRAM section about the benefits of HBM expansion to our equipment, our equipment is also benefiting from AI GPU and advanced packaging through chipletization. This page shows a schematic of the so-called chiplet packaging for HBM and AI GPUs, in which our deposition equipment is also used around capacitors in deep trenches within silicon interposers.
At present, deposition equipment is still being used in limited applications within interposers, but we expect sales of related equipment to continue to grow in this area, which is expected to receive a tailwind from advanced packaging. We are aiming to expand sales not only in leading-edge logic, but also in mature node logic. Until now, we have had a high market share in advanced processes, but we have lagged behind our competitors in mature nodes. However, we are now trying to turn things around by expanding sales of stable mature nodes, which will lead to broadening the base of our install base. Since many mature nodes are conventional processes, we are trying to differentiate ourselves from our competitors by utilizing the know-how we have cultivated in advanced nodes, such as productivity and energy-saving technologies.
While we have sold equipment from mature nodes in Japan and Asia in the past, we have now formed a strong local team in Europe and the U.S. and continue to sell to major mature node customers in Europe, the U.S., and China. Fortunately, inquiries have been strong, and we have already completed the startup of the evaluation equipment and expect to see the results in the next fiscal year. Finally, let me explain about treatments. Treatment systems, the second pillar of our business after batch ALD, continues to grow, driven by the need for film deposition in low-temperature environments and the increasing complexity of devices. Complex processes require deposition at low temperatures, while low-temperature deposition can result in insufficient film performance. Our treatment system, MARORA, can be used over a wide temperature range to improve film quality and further increase step coverage.
To date, the treatment has been mainly used in 3D NAND and has been adopted by several customers, with a high sales contribution, especially when NAND investment is active. We are now seeing more DRAM applications, and as explained earlier on the DRAM page, our major customers have already adopted our batch ALD and treatment combination for their HBM-related DRAMs. We are also expecting POR from other DRAM customers in the near future, which is also expected to be utilized for HBM-related applications. As a result, as shown on the right, we expect our treatment systems to grow at a CAGR of over 50% as the NAND market recovers and adoption in DRAM continues. We expect it will take more time for logic to be adopted, but several major customers are evaluating, and we believe there is a great opportunity here as well.
Now that we have explained the equipment, let's move on to services.
I am Yamamine. I am in charge of the service. This section describes the service strategy. Our service business, an important growth driver, has been steadily expanding the sales regardless of changes in market conditions, providing a stable base for our business performance. As shown in the upper left, parts and maintenance sales have achieved stable sa les growth even in volatile market conditions. The services we provide are highly valued by our customers, and we are confident that stable service sales will continue in the future. Our service business aims to grow along with the increase in the installed base and an increase in service sales per unit. As shown in the lower left graph, the installed base is growing steadily, and the parts and maintenance sales per unit are also increasing.
As shown in the lower right graph, service sales per unit are about four times higher for leading-edge equipment, and the more leading-edge equipments we sell, the higher service sales we can expect. We have service locations all over the world, and we are actively expanding new locations, especially as the semiconductor supply chain is becoming more localized amidst recent geopolitical developments. We are aiming to further expand our service business by establishing new bases in regions where fabs are being constructed, such as Dresden, Germany, Kaohsiung, Taiwan, and Hokkaido, Japan. In addition, we have recently established a Singapore office as a subsidiary rather than a branch to prepare for further growth in Southeast Asia, India, and other regions. The service business includes sales of 150 mm and 200 mm equipment, of which equipments for SiC power devices have achieved high growth.
Sales of conventional equipment are currently expanding, and sales for SiC power devices are expected to grow from JPY 500 million to JPY 4 billion between last year and this fiscal year. In the next fiscal year, we plan to sell high-temperature annealing systems for advanced processes, which is expected to further contribute to sales. As shown in the right of the chart, we are steadily building up PORs for SiC customers and have been able to acquire new PORs for 200 mm in addition to 150 mm. The strength of our equipment for SiC power devices is shown here. First of all, the strength of our equipment is that we can handle almost all processes using a common platform called the VERTRON Revolution, leveraging our expertise in cutting-edge thermal technology.
This allows our customers to save maintenance member hours, increase inventory efficiency, and improve productivity, which is why our equipment is chosen by customers. In addition, SiC substrates are much more expensive than silicon substrates, and the system to protect the wafer provided by our equipment has been well received. For SiC power devices, we anticipate the mass production and sales of annealing equipment for high temperatures in the next fiscal year, and we plan to achieve further growth by expanding sales of equipment for advanced processes using ALD technology.
So I am Yamada. I am in charge of production and procurement. This section describes our global production system, procurement, and operations. First, we are talking about the global production capacity expansion plan. We currently have two production bases, a main Toyama plant in Japan and Cheonan plant in Korea.
In order to expand production capacity in advance of growth of the WFE market, we are constructing a new plant as a new production base in Tonami City, Toyama Prefecture. When the new plant is completed and in operation, as the graph shows, the global production capacity will almost double from the fiscal year ending March 2021 to the fiscal year ending March 2026. We believe that this production capacity will be sufficient to meet the long-term growth of the WFE market. This slide shows the outline and project concept of Tonami plant. As we have already disclosed, Tonami plant is a 40,000 square meter site, three stories above ground, and is being constructed with an investment of around JPY 24 billion. Currently, construction and operation plans are progressing smoothly, with operations scheduled to begin around October.
The new plant will be a smart factory and will promote activities based on the SFX200 project concept, which aims to more than double production capacity as well as production efficiency. To achieve this goal, we will systematically promote smarter production processes, smarter material handling, and smarter facility management. The facility will also operate on 100% renewable energy in consideration of the environment and will be equipped with BCP function in preparation for natural disasters. This slide shows a brief summary of our initiatives to improve production efficiency. In order to double production efficiency, the following two initiatives will be promoted. The first is to introduce a new production system with a modularized production system for efficient operation using less space.
Second, in order to increase the turnover rate in the production area, we are working on smarter production processes and smarter material handling systems that carry items synchronized with the process. This will shorten the lead time. To make our operations smarter, we plan to systematically introduce cutting-edge technologies, including IT, IoT, digitalization, data utilization, automation, and even AI. This slide is an explanatory page regarding the environmental considerations of the new office and the business continuity at the BCP in the event of disasters. All electricity to run the plant will be generated from renewable energy sources. As shown in the figure, the rooftop, as well as the parking lot, will feature a new solar power generation system. We will also introduce an energy management system, EMS, that efficiently manages energy generated to save energy.
In addition, the entire building is equipped with seismic isolation equipment as a BCP function in case of disaster, and the construction land has been raised 2 meters in anticipation of river flooding and flooding. In addition, a storage battery system will be installed in case of other power outages, and a water recycling and circulation system will be added in case of water outages to enable safe, secure, and stable production activities. From here, I would like to introduce our procurement initiatives. This slide shows our procurement efforts to respond flexibly to changes in the WFE market. In order to build a strong procurement system, we will classify procurement items and businesses based on the assumptions of short, medium, and long-term changes in the market. We are organizing a three-dimensional cube strategy.
Based on this organization, we will work with each partner in the supply chain, utilizing portfolio management by items. We have established a partnering policy and are promoting these activities. Through these activities, we are expanding the scale of procurement, securing stable supplies, and addressing cost competitiveness. In this slide, we introduce activities toward a sustainable procurement supply chain. The purpose is to share the group's philosophy and the policies and comply with global norms and collaboratively manage labor, safety, environment, quality, and compliance and BCP. Our systematized philosophy, policies, guidelines, etc. are publicly shared on our own supply website. We also set up annual, quarterly, and monthly dialogue opportunities with our business partners to strengthen transparent and fair engagement.
In addition, in the activities of RBA, which is expected to have positive effects on CSR through the supply chain, we received the highest rating of Platinum status in the RBA audit in May this year. This is a presentation about our production and procurement activities. Thank you very much.
I'm Kawakami, overseeing finance as well as accounting. I'm the Senior VP. Allow me to discuss financial figures. Page 50 summarizes the midterm goals. WFE is expected to exceed $120 billion in the next three to four years, and we are targeting sales revenue of at least JPY 330 billion and an adjusted operating margin of at least 30%.
The medium to longer-term targets announced last September were based on the assumption that WFE will be $110 billion-$120 billion, and we are aiming for sales revenue of $300 billion-$330 billion and an adjusted operating margin of 28%-30%. Therefore, the new medium-term target is aimed to exceed the upper limit of the previous target while maintaining the same timeframe. For your reference, we have set ROE and ROIC targets to improve return on capital from a medium to longer-term perspective while remaining conscious of the cost of capital. We recognize that weighted average cost of capital, WACC, for the fiscal year ending March 2024 will be between 9%-10%, and our medium-term targets are to achieve at least twice that level, ROE of 25% or more, and ROIC of more than 23%. Page 51 summarizes the capital allocation policy.
As in the past, our top priority is investment for growth. Excluding large capital investments such as the construction of a new plant in Toyama Prefecture and the expansion of our demonstration room in Korea, we expect to make regular capital investments of JPY 4 billion-JPY 6 billion per year. Strategic alliances, including M&A, will also be considered, focusing on areas where significant synergies can be expected. As for shareholder returns, we plan to maintain a dividend payout ratio of 20%-30% based on adjusted net income as in the past. In addition, once net cash becomes positive, we plan to flexibly consider share buybacks in order to return approximately 70% of free cash flow after repayment of interest-bearing debt to shareholders. The total return ratio, including dividends and share buybacks, is expected to be approximately 50% when the medium-term target is achieved.
Page 52 shows the change in orders received, order backlog, and revenue from sales. Since we have a mix of orders with long lead times that go across different fiscal years and orders with short lead times that are booked as sales within the same quarter, and since the composition of these orders varies from quarter to quarter, they are not necessarily leading indicators of sales and earnings. For this reason, we refrain from disclosing information on a quarterly basis. However, we present here the semi-annual changes in order backlogs in order to provide an understanding of major changes and trends.
Orders for long lead time projects, which had increased due to supply chain disruptions in the fiscal year ended March 2022, have settled down, and the decline in demand that began in the second half of fiscal year ending March 2023 bottomed out in the first half of the fiscal year ending March 2024. As for order backlogs, long delivery, we expect shift to recover in the second half of fiscal year ending March 2024 and strong order intake recovering in the second half of fiscal year ending March 2025. As for order backlogs, long delivery projects are being converted to sales. 90% of the approximately JPY 150 billion order backlog at the end of fiscal year ending March 31, 2024, is expected to be converted to sales and normalized in the fiscal year ending March 2025.
As explained at the full year results earnings, sales and earnings bottomed out in the first half of the fiscal year ending March 2024 and have been on a recovery trend since the second half of fiscal year ending March 2024. We expect that mass production advanced devices will begin to recover globally in the second half of this fiscal year. From page 53 and onward, our main financial indicators. So we will not go into details. Thank you for your attention.