Enterprise productivity and Artificial Intelligence
Enterprise transformation with technology has limitless possibilities. Yet there is a large gap between ambition and execution in most enterprises. Most current processes are primarily human driven and supported by machines. Fast forward to the future, and most enterprises are predicted to run on AI and supported by humans. While the future of the likes of what is seen in the film Ex Machina is still a long way off, companies have started to use AI to drive autonomous processes and significantly improve operations and process agility.
According to the March 2018 IDC report, worldwide spending on AI and cognitive systems will increase by 52.7% in 2018 over 2017. IDC also predicted a phenomenal 46.2% annual growth rate of AI spending over the 2016-2021 forecast period. The most common use cases are forecasted to be around the areas of automated customer service, automated threat intelligence and prevention systems and sales process recommendation and automation.
Almost every industry is expected to be transformed with AI and Automation. In the manufacturing industry, AI is beginning to touch every stage of the supply chain to improve output including logistics, shop floor activities or maintenance processes. Several enterprises are automating their manual drudge work to focus on providing superior stakeholder experience, drive personalization and innovation. AI is driving customer experience in industries like retail and hospitality by understanding the context and providing contextually relevant solutions. In the energy and mining segment, AI could be applied to drilling, equipment maintenance as well tracking of tankers for instance. In the healthcare space, it could be used to predict diseases, automate diagnostic tests and increase accuracy of treatment. In the banking segment, AI can analyse contract documents and extract important clauses, leverage API’s to connect customers with best of breed providers, identify clients best positioned for cross sell and upsell, customer support, eradicate frauds, and predictive analytics.
Here are five ways that AI is being leveraged to drive enterprise productivity in enterprises:
Increasing the accuracy of demand forecasting:
Demand forecasting is typically extremely complex, entailing adaption to constantly changing variables amidst processing vast volumes of data, hundreds of mathematical models of production and outcome possibilities. With AI, demand and financial planners can extract knowledge from massive data sets from internal and external sources to provide accurate insights. Hundreds of advanced models could be simultaneously tested to generate more accurate forecasts, recommendations and predictions. What is interesting is that with AI, real time changes in information including competition, new channels, products etc. could be accounted for to ensure real time course correction and ensure that opportunities are leveraged in a timely manner. Demand and financial planners can spend less time on manual work and more time to make value added strategic decisions.
Predictive Maintenance, Predictive Insights:
Predictive maintenance is not new. Sectors such as Oil & Gas have been using technologies to forecast outages and improve maintenance efforts for more than a decade. Machine learning brings about a completely unprecedented level of accuracy and productivity at a large scale across several asset bases.
Traditional technologies are very data scientist labour intensive. With advanced AI algorithms, several classification and regression models are built across different classes of assets and data streams, the output of which generates a combination of results. To do this, requires a fundamental change in terms of breaking down internal boundaries between several departments and siloed enterprise IT systems. Almost 90% of business cases for IoT is around predictive maintenance. There is still a huge opportunity in organizations with big plants and/or large asset bases to significantly bring down costs with AI enabled predictive maintenance.
AI in general is capable to transform every function with much lower cost to make accurate predictions.
Procurement:
The procurement organization is redefining its value proposition because of technologies such as RPA and AI. AI can significantly improve procurement efficiencies such as spend analysis and contract analysis. With AI for example, past experiences with suppliers and current data could be used to re-write contract terms. In the strategic sourcing function, AI can learn from sourcing patterns and behaviours to make supplier recommendations, identify potential vendors and predict market prices. Anomalies in data sets such as unusual order frequencies and discrepancies in costs could be easily identified, to then send proactive alerts. Day to day tasks such as checking invoices against contracts and purchase orders, record keeping etc. can automate comparisons between thousands of lines of invoices and then automatically populates databases. Once these processes are automated, AI can integrate the functions of predictive analytics to understand buyer behaviours and adapt sourcing strategies according to the changing demand. AI is reported to reduce supply chain forecasting errors by over 50% and significantly reduce the costs relating to warehousing, sourcing and supply chain.
Improving customer interactions and customer service
With AI Chatbots, customer enquiries can be answered quickly, cheaply and consistently. Irrespective of the time in the day, customer responses can be tackled, as chatbots don’t sleep. Several volumes of similar requests can be tackled as chatbots don’t need time to search answers. The bot could strike up friendly conversations with customers and provide the most relevant information for them based on understanding their contextual requirement. The bot could also inform customers about promotions, based on their buying pattern and offer relevant product pages and images to generate more interest. It could perform simple transactions such as booking tickets, hotel rooms etc.
The chatbot could direct questions to a human only when it cannot settle the matter. This is expected to reduce service time by five-fold and significantly improve employee productivity. Sharing customer intelligence with agents enables them to have prior information with respect to the intent of the call and well as customer data. This allows for faster resolution and better customer experience.
Intelligent Personal Assistant
Employees typically work with multiple applications and access different types of data to get their work completed or to make decisions. This is typically quite tedious and labour intensive. With AI enabled chatbots, employees could now obtain data at their fingertips without opening new applications or windows. The device’s microphone could receive voice requests while the voice output takes place at the speaker. The presentation of such data could be in an easily consumable format that enables better productivity and process agility. Chatbots could be designed to push updates and recommendations to automate internal communication, prioritizing emails, preventing sharing of confidential information, balancing workflows, sending reminders etc., ultimately making lives easier for employees. The Intelligent Personal Assistant just allows employees to be more mobile, work smarter and improve employee morale.
Apart from the above, AI has limitless use cases to improve productivity including automation of time tracking, predicting areas for sales visits, forecasting costs for reimbursements for such sales visits, employee onboarding processes, reduce frauds, bad debts, automatically start a private conversation in response to comments from a social post, automate customer interactions and routine human assistant tasks.
According to Forrester, “Companies that master the interplay between AI, automation and human relationships will dominate their industries”. The success of AI is however dependent on human interaction to train the AI and ensure that technology does the job in a consistent way that mimics human intelligence. Again, the more data that is fed, the better the outcome with AI. Having large data sets is fundamental to any AI project success. It is also important to put AI to work where there is maximum impact on revenues and costs. Unlike popular perception that AI should be first deployed for the front office, according to a recent TCS research, 51% of AI leaders predicted that by 2020, AI will have its biggest internal impact on back-office functions such as finance and IT.
Humanization of technology is being tried and tested for several decades to varying degrees of success. While the emotional intelligence of machines goes up, end of the day we live in a complex multi-dimensional world that may not be only restricted to interpretations with algorithms. To maximize returns from AI investments, it is paramount to define an AI strategy that includes identifying processes that can give quick returns by leveraging this technology. Indeed, when done right, implementing this technology will help enterprises differentiate themselves via much more agile, borderless processes and better stakeholder experiences.
