Analysis on the optimum deployment strategy for quantum computer

Quantum computer is an outstanding innovation, offering possibility in solving complicated computational problem that current supercomputers are not able to solve. Experimental development of quantum computer has shown tremendous progress in the past decades. However, the implementation of quantum computer is not yet roaring. Therefore, this study examines the optimum deployment strategy for quantum computer. The study focuses on three main elements in the deployment process, which are: launching time, price and distribution. The suggested deployment strategy for quantum computer is: (1) avoiding subcontractors by investing internally on production of quantum components, (2) applying high price strategy, and (3) applying direct-to-customer distribution model in combination with pull marketing strategy.

Introduction

In the past decade, there has been tremendous progress in the experimental development of a quantum computer: a machine that would exploit the full complexity of a many-particle quantum wavefunction to solve a computational problem. A quantum computer will not be a faster, bigger or smaller version of an ordinary computer. Rather, it will be a different kind of computer, engineered to control coherent quantum mechanical waves for different applications. The example task for quantum computers which has provided the foremost motivation for their development is Shor’s quantum algorithm for factoring large numbers (Ladd et al., 2010). This is one among several quantum algorithms that would allow modestly sized quantum computers to outperform the largest classical supercomputers in solving some specific problems important for data encryption. Therefore, quantum computer becomes an outstanding innovation for the computing sector. 

As quantum computer promises robust functionality and shows a positive development progress, implementation of this innovation is of interest. Deployment is core part of the innovation process itself. Deployment strategies can influence the receptivity of customers, distributors, and complementary goods providers. An effective deployment strategy can reduce uncertainty about the product, lower resistance to switching from competing or substitute goods, and accelerating adoption. In contrast, ineffective deployment strategy can cause brilliant technological innovation to fail. Therefore, it is necessary to analyse and design an optimum deployment strategy for quantum computer.

This study analyses the optimum deployment strategy for quantum computer by considering maturity of the quantum technology and relating it with the customers demand and deployment process. The study focuses on three main elements of deployment process, which are: launching time, pricing and distribution. Section two examines the progress of quantum components and its production challenges, while section three discusses about the prospective users of quantum computers from design thinking perspective. Finally, formulation of the deployment strategy is addressed in section four. 

Maturity of quantum computers

The history of quantum computing began with the academic question concerning the minimum amount of heat produced in one computational step. In the early 1980s, Benioff and Feynman introduced the idea of quantum computers for the first time (Vogel, 2011). They showed that bits represented by quantum-mechanical states can evolve under the action of quantum-mechanical operators to provide reversible computation. Nowadays, the development of quantum computers is experiencing a major boost. In November 2017, IBM built and tested a 50-qubits processor, while in March 2018, Google announced a 72-qubits processor. Intel and Alibaba are also working on double-digit-qubits proof-of-concepts (Bova, Goldfarb and Melko, 2021).

Development of the quantum computers covers two related computational aspects, which are quantum algorithm and hardware. In 1999, Shor discovered quantum the first algorithm that gave a substantial speedup over a classical algorithm for a well-studied and interesting problem (Shor, 1999). Since then, research on quantum algorithm has rapidly grown as a hope for other interesting quantum algorithms risen. Fourier transform based quantum algorithms, Grover’s algorithm, and simulating quantum mechanics are the three candidates for a promising versatility and powerful quantum algorithm (Grover, 1997). However, the discovered quantum algorithms are not yet outperforming the classical algorithms(Shor, 2005). Two possible reasons are; first is only a few problems for which quantum computers can offer a substantial speed-up over classical computer. The second is that quantum computers operate in a different manner compared to classical computers, so that the experience gained for the classical algorithms offers little insight into building the quantum algorithms. Consequently, although quantum computers are believed to deliver faster and more efficient computing than classical computers, these limitations can be a challenge for research and development of quantum computers itself. 

The most critical aspect of quantum computers is the ‘closed box’ requirements. A quantum computer’s internal operation must be isolated from the rest of the Universe because small amounts of information leakage can disturb the system (Ladd et al., 2010; Vogel, 2011). Hardware for quantum computers should thus be able to maintain the simultaneous quantum systems, measure them, and preserve the isolation. Breakthrough in quantum hardware, such as Knill-Laflamme-Milburn scheme (Knill, Laflamme and Milburn, 2001), trapped atoms (Bruzewicz et al., 2019), superconducting qubits (Aguado, 2020), has been reported in the past decades following the development of quantum algorithms. This indicates that there are options and possibilities for achieving a large-scale quantum computers. However, two main challenges for the quantum hardware were identified; first, relatively few material platforms have been explored thus far, and second, there are several materials issues that do not affect single-qubit operations but appear as limitation for scaling up to a larger system (de Leon et al, 2021).

In general, development of quantum computers shows a positive progress in the past decades where it can be considered as a well-developed technology. Although a large-scale quantum computer is facing some challenges related to materials selections and algorithms, prototyping and building a small-scale quantum computer for a specific purpose are still feasible.

Consumers demand on quantum computers

Many leading businesses have no longer seen developing the same goods and services as enough to be success in the highly competitive global market. As a result, the current state of business innovation focuses on creating user experiences and developing system for living, working, and entertaining (Muratovski, 2015). Design and business are thus intrinsically linked. The field of design evolves as the business models began to evolve where designers have shifted from being stylists to becoming professional problem solvers. Design thinking is the best tool so far for problem-solving-based business innovation, which involves empathy with users, prototyping and tolerance for failure (Kolko, 2015). The launch of quantum computers must be then related to the current users’ problems, particularly how quantum computers can solve users’ problems.

Quantum computers promise their potential to drastically reduce the time to solve combinatorics problems by utilizing algorithms that make use of quantum effects. Combinatorics problems lay on finding ways to combine a set of objects, such as whether a certain combination is possible or what combination of objects are “best” by some metric. In many cases, the number of tasks required to answer these questions grows exponentially as the number object grows. This makes finding the answer computationally challenging. 

Combinatorics challenges are common in banking and finance, from arbitrage to credit scoring to derivatives development (Lopez de Prado, 2015). Many of these challenges relate to the classic “travelling salesman problem” that has been a staple of operations research for decades (R.M., 1972). The idea of the travelling salesman problem is that one salesman has a number of cities to travel to and needs to travel to every city once. The goal is to find the shortest route that goes to each city once and ends up in the starting city. From the value proportion perspective, the benefit to using the shortest route is straightforward; the salesman will presumably generate the same revenue by travelling to each city, while minimizing travel costs by pursuing the most efficient route. It is relatively straightforward to find the shortest route when there are comparatively few cities to travel to. However, the problem becomes less and less traceable when more cities are added. Quantum computers can offer the potential to speed up the process of finding the shortest route up to the point that finding a global minimum route through all possible cities becomes possible. 

In advanced manufacturing sector, the main problem lays on finding the sequence of events that led to failure among many possible sequences, which is similar to the travelling salesman problem. For instance, in the process of making CPU chips, there may be 2 failures in 10,000 runs once the machines are properly calibrated. Additionally, there can be thousands of steps in the process to manufacture a chip, and every step in the process may have different sensors and indicators that can take on different readings. In this settings, the number of combinations that need to be assessed increases exponentially with each new process step. If failure is common (e.g. 20% failure rate), then conventional statistical methods to inferring causal links between various factors and failure would become viable. Furthermore, for less advanced manufacturing, such as clothing manufacturing which can involve as few as 15 steps, standard solutions are also more feasible because the combinatorics problems are not as challenging. Although solving the problems gets more challenging with each step, there are materially fewer steps in this process and thus far fewer combinations to assess. Therefore, quantum computers are not as likely to provide a benefit beyond current methods in settings with high failure rates or in less advanced manufacturing with fewer steps. 

In short, quantum computers offer short computing time to solve combinatorics problems where sufficiently high number of objects involve in the process. This makes finding global minimum route or a sequence of events that led to product failure being feasible to perform on surprisingly large number and unbalanced data set. However, quantum computers may not provide better solution than the current computing methods for less complicated problems and data set. Therefore, these indicate that large firms, such as large-scale businesses or governments, will become the main users of quantum computers.

Deployment strategy for quantum computers

LAUNCHING TIME

According to Schilling, the optimal timing of entry is a function of several factors: (1) the margin advantage offered by the new innovation, (2) the state of enabling technologies and complements, (3) the state of customer expectations, (4) the threat of competitive entry, whether the industry faces increasing returns and a firm’s resources (Schilling, 2020). In addition to short computing time for complex and unbalanced data set, quantum computers can solve complicated problem which current computing technology are not able to perform. This shows that the margin advantage of quantum computers is relatively high. Development of hardware and algorithm for quantum computers have shown positive progress in the past decades where building a small-scale quantum computer for a specific purpose is currently feasible. Banking, finance, and advanced manufacturing sectors have expressed their needs in quantum computers for solving problems, such as finding the sequence of events that led to failure among many possible sequences. Since there are no many options for the future computing technology, quantum computers appear to be the most promising innovation within computing sectors. These say that quantum computers fulfil Schilling’s launching prerequisites and thus are ready for launching. 

Although quantum computer seems ready to launch, there are few aspects that need to be considered and learnt from the Boeing’s Dreamliner project as both cases may undergo similar issue in terms of new innovation (Shenhar et al., 2016). Boeing aimed to make Dreamliner as the most advance and efficient commercial aircraft ever built. However, its late delivery and early service problems were particularly troubling for Boeing. Boeing’s difficulties were apparently a result of major challenges, such as the use of newly developed technologies, outsourcing a large extent of design to less experienced subcontractors and creating development chain, a new business model of revenue sharing and a new assembly model. Quantum hardware and algorithms are just recently developed for small-scale devices while development of large-scale devices is still on going. Similar to the Dreamliner project, developers of quantum computer will most likely face less experience subcontractors in terms of hardware components and software developer. In this case, the quantum computer developer should invest on production and development of quantum components internally, instead of outsourcing it to the subcontractors (McIvor, 2008). Internal investment on the production and development of quantum components is a big economic task for a company. However, as quantum computers offer potential for future growth, investing on production and development of quantum components may exhibit advantages in the future, such as increase of company’s competitiveness. To simplify the development process of quantum components, developers of quantum computer might need Agile-State-Gate method applied to their management strategy. Agile-State-Gate adds the most value when there is high uncertainty and a great need for experimentation and failing fast (Cooper and Sommer, 2016).  The combination of avoiding outsourcing through internal investment and applying Agile-State-Gate as management strategy in the development chain is an optimum strategy for quantum computer developers in response to the Boeing’s Dreamliner project case. 

Schilling covers basic function for the optimum entry timing of an innovation. However, there can be additional factors applied to a new innovation where highly complicated project is involved. These additional functions are: (1) availability of the company to invest internally for production of small-critical components and (2) flexibility of the company in adjusting their engineering management strategy. In the case of quantum computers, the optimum launching time is when developers are able to invest internally for production and development of quantum components, and being ready to apply Agile-State-Gate method as their management strategy in the development chain. Otherwise, quantum computer developers may undergo some difficulties after launching such as late delivery and early service problems.

PRICING AND DISTRIBUTION

Pricing is a crucial element in the deployment strategy. Price simultaneously influences product’s positioning in the market place, rate of adoption, and the company’s cash flow. Before a company can determine its pricing strategy, it must determine the objectives it has for its pricing model (Schilling, 2020). In the case of quantum computers, developers need to consider two aspects for their pricing model, which are: (1) covering the return of investment for production and development of quantum components and (2) elongating the product’s adoption time for allowing technology to develop. High price strategy is a common pricing strategy where company sets a high price on their new products (Dolgui and Proth, 2010). The high price may signal the market that the new product is a significant innovation which offers a substantial performance improvement over previously available products. This strategy is also profitable for the companies. However, this pricing strategy may slow down the adoption time of the product. This is a benefit for the quantum computer. Since large-scale quantum computers are under development, slowing down the product adoption will provide more time for developers to scale-up the technology. Therefore, setting high price strategy is appropriate for quantum computer developers.

Companies can sell their products directly to users through their direct sales force or an online ordering system or mail-order catalogue. Alternatively, companies can use intermediaries such as manufacturer’s representatives, wholesales, and retailers. Selling direct gives the companies more control over the selling process, pricing, and service. However, selling direct can also be impractical or overly expensive in some situations. At the current state of quantum computers, large-scale businesses, organisations and governmental institutions are the main users and consumer of the quantum products. These large organisations often require demand for customised products that are suitable for their special case and needs. Quantum computer developers should thus expect an intense sales-customer interaction and consumer-centric marketing strategy. Accordingly, direct-to-customer distribution model in combination with pull marketing strategy becomes a good distribution strategy for deploying quantum computers. 

Conclusion

Development of quantum computers has been showing a positive progress for the past decades where it can be considered as a well-developed technology. However, production of quantum computer is still limited to prototyping and small-scale devices for specific purpose due to limited materials selections and algorithms. Consequently, the users of quantum computers are limited to large-scale businesses, organisations, and governmental institutions. Hence, the suggested entry strategy for quantum computers is as follow:

1. Product launching should be done after the developers of quantum computer are willing to invest internally for production and development of critical quantum components and being able to adjust their engineering management strategy. Otherwise, developers may undergo difficulties after launching, such as late delivery and early service problem.

2. Pricing model for quantum computers should consider covering the return of investment for production and development of quantum components and elongating the product’s adoption time for allowing technology to develop. High price strategy is considered as appropriate to this case.

3. Since the main consumers of quantum computers often demand for customised products, quantum computer developers should consider direct-to-customer distribution model in combination with pull marketing strategy as their distribution strategy.

References

Aguado, R. (2020) ‘A perspective on semiconductor-based superconducting qubits’, Applied Physics Letters, 117(24), p. 240501. doi: 10.1063/5.0024124.

Bova, F., Goldfarb, A. and Melko, R. G. (2021) ‘Commercial applications of quantum computing’, EPJ Quantum Technology. The Author(s), 8(1), pp. 1–13. doi: 10.1140/epjqt/s40507-021-00091-1.

Bruzewicz, C. D. et al. (2019) ‘Trapped-ion quantum computing: Progress and challenges’, Applied Physics Reviews. AIP Publishing LLC, 6(2). doi: 10.1063/1.5088164.

Cooper, R. G. and Sommer, A. F. (2016) ‘From experience: the Agile-State-Gate hybrid model: a promising new approach and a new research opportunity’, J PROD INNOV MANAG. doi: DOI: 10.1111/jpim.12314.

Dolgui, A. and Proth, J.-M. (2010) ‘Pricing strategies and models’, Annual Reviews in Control, 34(1), pp. 101–110. doi: https://doi.org/10.1016/j.arcontrol.2010.02.005.

Grover, L. K. (1997) ‘Quantum Mechanics Helps in Searching for a Needle in a Haystack’, Phys. Rev. Lett. American Physical Society, 79(2), pp. 325–328. doi: 10.1103/PhysRevLett.79.325.

Knill, E., Laflamme, R. and Milburn, G. J. (2001) ‘A scheme for efficient quantum computation with linear optics’, Nature, 409(6816), pp. 46–52. doi: 10.1038/35051009.

Kolko, J. (2015) ‘Design thinking comes of age’, Harvard Business Review, 2015(September).

Ladd, T. D. et al. (2010) ‘Quantum computers’, Nature. Nature Publishing Group, 464(7285), pp. 45–53. doi: 10.1038/nature08812.

de Leon, N. P. et al. (2021) ‘Materials challenges and opportunities for quantum computing hardware’, Science, 372(6539), p. eabb2823. doi: 10.1126/science.abb2823.

Lopez de Prado, M. (2015) ‘Generalized Optimal Trading Trajectories: A Financial Quantum Computing Application’, SSRN Electronic Journal, pp. 1–11. doi: 10.2139/ssrn.2575184.

McIvor, R. (2008) ‘What is the right outsourcing strategy for your process?’, European Management Journal, 26(1), pp. 24–34. doi: 10.1016/j.emj.2007.08.008.

Muratovski, G. (2015) ‘Paradigm Shift: Report on the New Role of Design in Business and Society’, She Ji. Elsevier Ltd, 1(2), pp. 118–139. doi: 10.1016/j.sheji.2015.11.002.

R.M., K. (1972) Reducibility among Combinatorial Problems. Springer, Boston, MA. doi: https://doi.org/10.1007/978-1-4684-2001-2_9.

Schilling, M. A. (2020) Strategic Management of Technological Innovation. 6th edn, McGraw-Hill. 6th edn.

Shenhar, A. J. et al. (2016) ‘The challenge of innovation in highly complex projects: what can we learn from Boeing’s Dreamliner experience?’, Project Management Journal. doi: 10.1002/pmj.

Shor, P. W. (1999) ‘Polynomial-Time Algorithms for Prime Factorization and Discrete Logarithms on a Quantum Computer’, SIAM Review, 41(2), pp. 303–332. doi: 10.1137/S0036144598347011.

Shor, P. W. (2005) ‘Progress in quantum algorithms’, Experimental Aspects of Quantum Computing, 3(October), pp. 5–13. doi: 10.1007/0-387-27732-3_2.

Vogel, M. (2011) Quantum Computation and Quantum Information, by M.A. Nielsen and I.L. Chuang, Contemporary Physics. doi: 10.1080/00107514.2011.587535.