Implementation of Artificial Intelligence at Woolworths Limited

Question: Discuss about the Implementation of Artificial Intelligence at Woolworths Limited. Answer: Introduction: Artificial intelligence has showcased exponential development in the past two decades capitalizing upon the development in computer science, operations research, nanotechnology and neuroscience. The convergence of several fields with an technology driven approach has presented with overwhelming possibilities in terms of benefits derived by end users along with high growth potential as an business opportunity. Woolworths Limited has the largest market share in retail sector of Australia with encapsulating nearly 80% of retail consumer base. Global retail giants such as Amazon tends to emphasize upon numerous sets consumer based researches in order to improvise upon the deliverables to be offered to its online visitors. The implementation of artificial intelligence (AI) at Woolworths will improvise upon the companys comparative advantage both in terms of retention of Australian retail market along with overseas consumer base. The case study seeks to highlight different sets of AI applications that can potentially benefit the company in terms of improving retail experience for its consumers. The current sets of applications deployed by the company focuses towards presenting products online to a consumer based upon his/her sales history. However, AI can add numerous dimensions towards such recommendations through undertaking sentimental analysis and shifts in trends to come up with relevant searches. Problems addressed by Artificial Intelligence: The online retail experience of consumers tends to get affected owing to a large number of product alternatives available along with different sets of variant in each product segment. The paradox of choice results in aggravating of confusion of prospective consumers about to get their initial sets of retail shopping. Moreover, the search engines are not fine tuned in order to provide search results based upon consumers preferences. Consumers seek assistance progression of the kind that are not fully addressed by the present sets of measures available at Woolworths online retail sites. The marketing campaigns undertaken by the company currently employs large number of employees entrusted with promotional activities or outreach programs for its consumers. Moreover, the prevalence of non alignment between the actual tangible form of the product and its online representation through photographs and product details results in high rate of dissatisfaction amongst aggrieved clients. This is owing to the fact that such non alignment results towards the products not reaching the initial benchmark that has been set up by the clients. In terms of search engine optimization process inside the Woolworthss online stores, the company has to improvise upon bringing out relevant searches based upon the search term entered by the consumer coupled with his/her browsing history. AI can improve the search processes through personalized searches for each consumer that equates with their previous transactions with Woolworths Limited. In case of brick and mortar stores by Woolworths, the dissemination and communication of different sets of offers provided by the company towards targeted consumers requires precision in terms maintaining databases. AI can have positive implications both in terms of augmenting the consumer experience and streamlining the degree of sustainable growth in revenue for the company through error free maintenance of different sets of company operations. Potential benefits that can be derived using Artificial Intelligence: From the initiation of product search in the retail website to resolving of issues pertaining to logistics and delivery of such products to its end users, AI can function upon numerous set of parameters. Moreover, in order to optimize the transactions made by the consumers and assist the company towards improving its revenue generating capabilities through elevating the degree of customer satisfaction, the role of AI tends to be crucial. Through deployment of machine learning algorithm based programs that evolves based upon the data sets made available to it, better sets of predictions as regards to changing consumers preferences and online activity can be undertaken. Wenger (2014) states that machine learning programs are not application specific and tends to modify itself in accordance with the data made available to it and analyses the data structure in order to decide upon the degree of optimization. Artificial intelligence, through Reinforcement Learning can also contribute towa rds upgrading the security profile of the online site of Woolworths through heightening the degree of surveillance on suspicious activities on its network. Figure 1: Machine Intelligence (Source: Celent.com, 2016) Online retailing encompasses dissemination of sensitive financial information while putting forward a transaction for billing purposes. Thereby, the bank and credit card details has a significant probability of getting compromised with the advent of sophisticated malware. Artificial intelligence optimizes the existing security to such an extent through multilevel encryptions that malware and unwarranted programs ceases to persist (Bond Gasser, Eds, 2014).Moreover, through constant evolving of data functions and detailing of operational and infrastructural problems based upon the consumer grievances, help desk incidents and computer logs AI can significantly reduce probability of potential threats. AI has the potential to replace human shopping consultant through customizations of product options based upon the consumers behavioral analytics, sensitivity towards the different sets of products offered and up-gradations regarding emerging consumption trends. In terms of recruitment of employees by Woolworths, AI can improvise upon short listing of candidates based upon pre-programmed metrics such as previous work experience, reasoning skills, experience in sales. Further, due to the quantum of training procedure involved for sales staff along with progressive sets of skill acquired by an employee, turnover costs remains high. Implementing AI in recruiting reduces the probability of employee turnover owing to optimizing the selection process (Seibt, Hakli Nrskov, 2014).. Supply chain management remains an issue even for the global leaders in retailing, through algorithmic SCM technologies using data from machine learning both the inventory management and distribution can be automated in tune with the sales forecasts. Further, the stock out rates along with backorders that tends to affect the consumer retention rate significantly can be minimized efficiently through AI based forecast modeling. The effectiveness of different sets of market ing strategy can be gauged with precise sets of event timings and keywords that were used for such marketing campaign using machine learning techniques . Implementation of data modeling procedures in order to compute sales and revenue returns from different sets of discount offerings and promotional events pertaining to Woolworths and its market competitors can be undertaken using predictive analytics tools. Figure 2: Life cycle of AI installation (Source: Bond, A. H., Gasser, L. Eds., 2014) In case of decision making with regards to opening up of stores, data regarding demographics, income distribution amongst the local population, proximity to community events and proximity to nearby competitors can be evaluated by AI algorithmic game theory and data modeling. The recent success of AlphaGo, an AI developed by Google, over a professional player in a five match strategy game showcases the fact that AI can excel through Reinforced Learning and even excel human in terms of strategizing skills (News Blog | DeepMind. 2016). Thereby, implementation and developing a separate AI division that caters to different sets of retail management, supply chain configuration and customer relationship management through Deep Learning, Reinforced Learning and Natural Language Processing is imperative for Woolworths. Strategic plan: The initiation of implementing AI based technology requires reconfiguration of online metrics to facilitate merging of AI based resources onto the site metrics. The initial alignment between AI and the online site requires installation of additional sets of servers that tends to accommodate addition sets of random access memory reconfiguration. The high growth prospects in artificial intelligence along with the applicability in terms of collecting and synthesizing behavioral analytics data results towards exponential augmentation of revenue generating capability. The investment scaling taking into consideration the demand and sales estimates can be implemented through AI applications that will seek to evaluate probable constraints to such escalations. Moreover, the scaling of information network security through implementing machine learning into the system servers will aggravate the detection rate of malware previously undetected into the system. Figure 3: AI installation into corporate lifecycle (Source: News Blog | DeepMind., 2016) The company should initially undertake installation of IBM Watson on to its networks, prior to which it requires to recalibrate its existing IT infrastructure. The installation of bandwidth extension has to be initiated for peripherals coupled with expansion of cloud based services in order to provide the basis augmentation. The recruitment of system engineers and computer science doctorates on to the AI team shall facilitate towards dissecting different the underpinnings of neutral networks that tends to operate the AI. Further, only cost minimization of existing facilities cannot be the sole source of covering installation costs. There ought to be a source of cash inflow in order to further capitalize upon the RD and AI expenditures that are to be incurred by the company. Figure 4: Global revenue forecasts from Artificial Intelligence (Source: Artificial Intelligence for Enterprise Applications | Tractica. 2016) Thereby, the company should outsource its cloud based services onto other minor companies that are technology driven but do not have the adequate infrastructure to utilize the latest developments in the field of Artificial Intelligence. For instance, the company may seek to provide AI services to the different sets of transportation and healthcare companies for which AI system can provided larger breakthroughs in terms of providing cutting edge healthcare facilities coupled with improvement in the success rate of surgeries followed up gradation of transport routes or streamlining the logistical system in order to facilitate better returns for both the companies. Moreover, the insights gained from such a large consumer bases that Woolworths presently encapsulates in the 14 different economies that the company operates in. The company, owing to high degree of market exposure, can improvise towards providing highly valuable behavioral, preference and consumption patterns to different se ts of consumables and beverage manufacturers. Further, such insights can also be valuable towards providing highly consumer driven AI inputs. The above table showcases the fact that Asia Pacific region- the region where the majority of Woolworths business operations are undertaken has been forecasted to generate highest revenues by providing AI services. Thereby, Woolworths can capitalize heavily upon its access to consumers decision making metrics in order to facilitate cash inflows into the project. Project Analysis: CoCComputation of Total Expenditurempu Particulars A$ ('00) IBM Watson Installation 20,000 Development of AI Prototype 65,000 Recruitment of Software Engineers for AI Division 20,000 Development of existing IT Infrastructure 35,000 Expansion of cloud computing services 18,000 Installation of machine learning peripherals 7,500 Total Expenses 165,500 Table 1: Computation of initial cash flow (Source: As created by the author) Computation of NPV Cash Flow Cumulative cash flow Disccunted cash flow Year 0 -165500 1 Year 1 42000 (123,500.00) 0.892857 37,500.00 Year 2 47000 (76,500.00) 0.797194 37,468.11 Year 3 53000 (23,500.00) 0.71178 37,724.35 Year 4 64000 40,500.00 0.635518 40,673.16 Year 5 72000 112,500.00 0.567427 40,854.73 194,220.36 NPV 28,720.36 Table 2: Computation of Net Present Value (Source: As created by the author) The overall table mainly helps in depicting the NPV of the project, which might be helpful in making adequate investment decisions. In addition, the NPV of the project is mainly derived at 28720.36, which mainly indicates the efficiency of the project. In this context, Seibt, Hakli Nrskov (2014) mentioned that with the help fop NPV companies are able to determine the current value of the future cash flows by eliminating the time gap. On the other hand, Glymour, Scheines Spirtes (2014) criticizes that NPV valuation mainly loses its friction during an economic crisis, where the income projected from a project could be reduced and hamper liquidity of the company. Furthermore, the discounting factor is mainly taken at 12%, which helps in depicting the adequate income, which helps in discounting the cash inflows generated from the project. In addition, the overall cumulative cash flow mainly presents that at the fourth year the investment money will be collected by the company. Furthermore, the main cash inflows are mainly projected from the data sales that will be conducted to the retailers regarding the purchase behavior of their customers. This software might help in generating higher predictive analysis for the retailers, which in turn might reduce the excess storage of inventory conducted by retailers. Conclusion The application of artificial intelligence in retailing operations can be multifaceted with the presence of different approaches presented through Machine Learning, Language Processing and Retention Learning. The different sets of approaches resolves issues pertaining to separate domains of the retail operations. The presence of large consumer base tends to bring newer sets of challenges in the realm of customer retention and CRM activities. The different sets of issues with regards to network security, security in terms of transactions and the issues pertaining to supply chain bottleneck followed by logistical delays can be mitigated through the process optimization activities undertaken by AI. Moreover, Machine Learning evolves over time in order to improvise upon the data sets provided in terms of consumer preferences that it could utilize over time to mitigate the different sets of activities. The company can provide the insights derived from continuous data up-gradation with regards to changing consumer preferences to facilitate better access to AI driven data that minor or small scale companies do not have the means to procure. Thereby, the cash inflow through sourcing of such insights lead towards resolving the issue pertaining to installation costs. The overall NPV of the project is mainly depicted at 2872036, which mainly helps in evaluating the viability of the project. In addition, with the help of NPV the company might effectively continue with the project and increase return from income. References: Artificial Intelligence for Enterprise Applications | Tractica. (2016). Tractica.com. Retrieved 26 September 2016, from https://www.tractica.com/research/artificial-intelligence-for-enterprise-applications/ Bond, A. H., Gasser, L. (Eds.). (2014).Readings in distributed artificial intelligence. Morgan Kaufmann. Brodie, M. L., Mylopoulos, J., Schmidt, J. W. (Eds.). (2012).On conceptual modelling: Perspectives from artificial intelligence, databases, and programming languages. Springer Science Business Media. Celent.com. (2016). Retrieved 26 September 2016, from https://celent.com/system/files/ai_buyside_0.gif Cohen, P. R., Feigenbaum, E. A. (Eds.). (2014).The handbook of artificial intelligence(Vol. 3). Butterworth-Heinemann. Faggella, D. (2016). Valuing the Artificial Intelligence Market, Graphs and Predictions for 2016 and Beyond. TechEmergence.com. Retrieved 26 September 2016, from https://techemergence.com/valuing-the-artificial-intelligence-market-2016-and-beyond/ Glymour, C., Scheines, R., Spirtes, P. (2014).Discovering causal structure: Artificial intelligence, philosophy of science, and statistical modeling. Academic Press. Helgason, H. P., Thrisson, K. R. (2012). Attention Capabilities for AI Systems. InICINCO (1)(pp. 281-286). Hovy, E., Navigli, R., Ponzetto, S. P. (2013). Collaboratively built semi-structured content and Artificial Intelligence: The story so far.Artificial Intelligence,194, 2-27. News Blog | DeepMind. (2016). DeepMind. Retrieved 26 September 2016, from https://deepmind.com/blog/ Seibt, J., Hakli, R., Nrskov, M. (2014). Frontiers in Artificial Intelligence and Applications. Sotala, K. (2015, April). Concept learning for safe autonomous AI. In1st International Workshop on AI and Ethics, held within the 29th AAAI Conference on Artificial Intelligence (AAAI-2015). Michalski, R. S., Carbonell, J. G., Mitchell, T. M. (Eds.). (2013).Machine learning: An artificial intelligence approach. Springer Science Business Media. Wenger, E. (2014).Artificial intelligence and tutoring systems: computational and cognitive approaches to the communication of knowledge. Morgan Kaufmann.