The Role of Managers in Algorithm Management
Write about the Trends in the Global Business Environment for Algorithm Management.
In this report the discussion is done on the topic that “Algorithms need managers too”. Algorithms are used to create programs to support people by making the work automatic. Algorithms are designed by the coders but they need a proper management system to work efficiently. The humans design algorithm in such a way that they take decision regarding what need to be done in the system. Algorithms help in decision making but they need to be managed by managers in regular time interval.
The concept of algorithm management occurred few years back, it is just a way to influence managers to use them for tracking status. It does not keep supervision of the actions but perform the pre-defined tasks. The job of mangers was to make predications to achieve maximum efficiency. Nowadays algorithm itself makes predications to achieve goals, but as they take their decisions on their own, they increase the chances of risks. Thus in these cases managers are needed to manage the algorithms by understanding the concern and taking appropriate actions. Considering an example of Uber, the fares of each ride and the changes in fare due to heavy traffic is done by the algorithms only. The managers do not heck the fares each time a ride is done, algorithms smartly work upon the design policies. While designing the algorithms managers need to track all the possible ways to check that it is working correctly (Boston college, 2017). Algorithms work upon the desired actions, as the rules are encoded by the managers. Another example could be Netflix, as the algorithms itself take actions and provide the winners of the competition. The algorithms itself deploys the result of particular action which may degrade the overall customer satisfaction. These kind of problems need to be managed by the supervisors (Pringle, 2016). Mangers need to understand such situations to avoid missteps to happen.
On the other hand, bugs in the algorithms also need manager’s supervisions. An algorithm stops working if any kind of bug hits the system. In that case the algorithm needs to be managed by the mangers (Hebbar, 2016). Algorithm sometimes leads to astray as most of the algorithm are depends on the automated call function. Algorithms are treated to behave in the same way as humans but they are different as in some cases they excludes that pats that need to be considered and behave in a wrong manner (Luca, Kleinberg and Mullainathan, 2016). To deal with such situations artificial intelligence will not work and some managers are needed.
Algorithms can predict the future with great efficiency but they cannot provide the reason of occurrence of an event. As algorithms can think in a single manner with the same perception regarding the objective, thus managers are needed to tackle the difficult situation. Whenever a problem occurs in the system, algorithm can resolve the issue but it tackles the problem in a similar manner as they peruse the issue with single minded objective and they don’t understand trade off (Contamin and Ellison, 2010). Algorithms work correctly if the right data inputs are used; if the data inputs are entered incorrectly the algorithm will not provide the correct solution (Haskett, 2010). In such cases supervision are needed by the managers so that algorithm provides proper result. Otherwise algorithm makes use of existing data to make predictions and provide significant result. The result could be right or it can be wrong, this entirely depends upon the predication assumed by the algorithm. It is true that algorithms are capable of making decisions but this no don’t eliminate the need of managers to manage the algorithm. Mangers are needed to control the connection between the cause and effects (Parikh and Boyd, 2014). Managers are needed to look at the algorithms as they make mistakes and cause consequences. Sometimes the decisions made by algorithm can cause impact the environment in negative way. Thus to ensure that result of algorithm will not harm the responsibility, accuracy and audibility need to be checked by the mangers (Deloitte,2016). Algorithms need managers as they good at predicating things but sometimes perform certain actions that create risk in the system.
Algorithms and Decision Making
Algorithms may create undesirable result sometimes otherwise they are very literal and perform all the operations correctly. Sometimes algorithms are treated as black box because they provide incomplete information and the output may be the garbage value. When using algorithms manager must be explicitly clear regarding the goals. A manager can refine the result by including soft goals in their algorithm. Mangers are needed to resolve the issue of myopia as occasionally algorithms suffer from myopia as they look with the motive to complete their job as soon as possible. If a manager is not present the issue may affect the quality or speed of the result.
The case study related to does algorithms need managers too, helps in knowing the success of algorithms. Mangers involve various evaluation measures to find out the ways to maintain the value required to solve the problem by perusing a development objective. Mangers aim is to boost the solution so that algorithm can work correctly (Harrell, 2016). Algorithms sometimes undergo failure due to many types of problem which is difficult to be discovered by algorithm.
On the other hand it can be stated that mangers should not worry about the decisions made by the algorithms, it is not possible always that they make a wrong judgment. The work according to the rules as the predications of data made by algorithms is better than humans. Aalgorithms focus on how to maximizee the profit quickly and avoiding the use of low-quality content. While they focuses on increasing the clicks by advertising the services, hence as the number of clicks increase the customer satisfaction also increases. The algorithms function on their own but they work according to the plan designed by the mangers. Managers need to understand about their limitations so that they do not cause any harm to the system. A manager before polishing the algorithm checks that algorithms exactly work according to the design told to them by managers and ensures that they disregard every other consideration which is not mentioned by the managers. Thus it can be started that mangers are somewhere required to assure that algorithms work correctly (Mullainathan, Kleinberg and Luca, 2017). Other than that, managers need to take care of the causes an event can cause.
Managers need to manage the algorithms as with the innovation of new technologies algorithm need to be updated on regular basis. Mangers re need to update the algorithm and do modification whenever there is a change in the situation. One challenge faced by using algorithm is that humanity and human judgment is lost as data is modelled directly according to the algorithm designed. The algorithm tries to fit the perception of an individual but they do not assure that all teh changes will be modified accordingly in that cases managers help is needed. According to the research it can be stated that algorithm are rigid and they do not take care of all the factors before making decisions. Thus it is not a feasible method to make all the possible decisions, thus supervisors are needed to emphasis on all the factors before taking decisions.
The Need for Supervision in Algorithm Management
Managers are needed to manage the risk associated with the algorithms as they safeguard the use of algorithm by accelerating the performance and creating the solution. It is also possible that sometimes algorithms are not designed in a proper manner and it is vulnerable to risk due to wrong logic or poor assumptions. Inappropriate modelling techniques may lead to errors as they identify spurious patterns that are not required to make any decisions. Sometimes incomplete or irreverent data can cause diverse issue thus managers need to take care all the data is up to date and complete. Thus it can be concluded that managers are needed to manage all the risks associated with the algorithms so that long term implications due to affect the finical operational and technological background of the system.
Algorithms are important part of business as with the advancement in technologies companies and consumers simply rely on algorithm to analyse the data and produce the correct result. They are helpful for every business as they decrease the manual work of the organisation and taking the actions that are better than humans. The use of algorithm in businesses has enhanced the work by making everything automated from financial reports to strategies of the company are designed using algorithms. Additionally algorithms are very fast and they solve the problems within seconds (Schaltegger and Wagner, 2017). Algorithms are used in real world to solve the issues, as every operation performed on the computer is due to the algorithms. Even the easiest application works on the behalf of algorithm behind the scenes these algorithm needs to be managed by the managers to make sure that there is no load on the hard drive. Algorithm has become an integral part of every business as they have made the system automatic. Apart from that they solve complex problems which are not possible to be done manually (Pathak, et. al, 2014). For every digital business algorithms have enhanced the speed and decreased the man labour by handling large amount of data and helping the business to grow. It is very important to implement the use of algorithm in every organisation as it accelerates the rate of technology changes. It increases the connectivity between business and the rapid technological changes (Lenormand, et. al, 2015). Algorithms are important for business as they can alone compete with the problems and they can also be used to provide security to the organisation.
In some scenario it is possible that algorithm do not work up to the design which may cause harm to the system. In such cases algorithm need to be managed by the supervisors. For example mostly all the social media companies rely on algorithm to showcase the information to users according to their interest but in some cases algorithms fails and make wrong digital decisions. In such cases human oversight is needed to balance the automated decisions made by algorithms (Ullman, 2015). Apart from that mostly all the business are dependent on algorithms to deliver specific results if they goes wrong it can cause huge loss to the organisation. Algorithms are directly related with the customer satisfaction as they provide good customer experience by making the work automatic. If the algorithm goes wrong or fails the speed will degrade which will affect customer satisfaction (Hillery, 2016). Poor working of algorithms result in exploiting efficiency and causes decline in customer satisfaction which in turn destroys the image of the brand. If an algorithm fails it will create lack of clarity and will damage the integration of data.
Limitations of Algorithms
Algorithms are step by step procedures that are depicted by every organisation to make the work easy. They take specific set of inputs to provide desired output. These are used to perform repeated tasks by simply following set of instructions. This is very helpful for an organisation to solve business problem by making specific decisions. Algorithm helps in examining the problem and providing the solution by making a proper decision. Algorithm also helps in making improvement by presenting the solution to reduce the chances of risk (Medhi and Ramasamy, 2017). An algorithm is needed for every business to understand the problem so that list of actions could be constructed to get the desired output. Mangers help in testing the program by validating the programs and testing that it is working properly. Thus every algorithm should be feasible and effective to produce the desired output (Burger and Burge, 2016).
Advantage of algorithm- It helps in identifying the mistakes that can be caused by human and provides step by step solution for every problem. It uses definite procedure to understand the problem and provide the solution. The algorithm should be definite that means the algorithm should give clear meaning and the duration of algorithm should be finite (Lee, Kusbit, Metsky and Dabbish, 2015). The algorithm should be effective that means it should follow a particular procedure. We are surrounded by use of algorithm in everyday life, as it is use to solve a specific task. Every step in the algorithm should be precise and clear. Additionally, every step should be completed in a finite time. Steps should not be repeated infinitely, and the desired result should be obtained as soon as the algorithm is terminated.
Managers play an important role in managing the performance of an organisation by checking the accuracy of algorithm. Managers use various techniques to check the robustness of the algorithm by preparing the model to check the entire data set by checking the dataset and performing cross validation to ensure that accuracy is maintained. Managers check the algorithms on all possible input values to get the correct result. The goal of manager is to check the algorithm against all the problems by testing the outcome and the performance (Nielsen, et. al, 2015). To maintain the accuracy of the algorithm managers need check all the input specified in the algorithm and apply proper restrictions and rules so that no inappropriate data is entered. If the input is defined in a corrective manner it will result the desired output (Ananny, 2016). It is important for the managers to check the effectiveness of the algorithms and assure that it would work in all the conditions and will not cause any potential loss to the business.
Conclusion
Thus it can be concluded from the above article that mangers are sometimes needed to manage the operations of algorithms. Once the algorithm is designed they need to be modified with the change in technology. Algorithms need managers too as there are certain situation in which algorithm do not produce correct result and cause harm to the company thus to deal with such scenarios mangers are needed. In many cases algorithms don’t work correctly sue to occurrence of bugs thus managers tackle such situation so that no harm is caused to the company. Apart from the article few questions have been discussed like how have algorithm impacted the growth of business and various concerns that occur if the algorithm are not working correctly. Additionally it has also been discussed that how mangers maintain the accuracy of the algorithms.
References
Ananny, M. (2016). Toward an ethics of algorithms: Convening, observation, probability, and timeliness. Science, Technology, & Human Values, 41(1), pp.93-117.
Boston college, (2017). ALGORITHMS NEED MANAGERS / THE CURIOUS CASE OF IPOD SHUFFLE. Available from https://hbr.org/2016/01/algorithms-need-managers-too Accessed on 25 April 2018.
Burger, W. and Burge, M.J. (2016). Digital image processing: an algorithmic introduction using Java. Springer.
Contamin, R. and Ellison, A.M. (2010). Indicators of regime shifts in ecological systems: what do we need to know and when do we need to know it. Ecological Applications, 19(3), pp.799-816.
Deloitte, (2016). Managing algorithmic risks. Available from https://www2.deloitte.com/us/en/pages/risk/articles/algorithmic-machine-learning-risk-management.html Accessed on 25 April 2018.
Harrell, E. (2016). Managers Shouldn’t Fear Algorithm-Based Decision Making. Available from https://hbr.org/2016/09/managers-shouldnt-fear-algorithm-based-decision-making Accessed on 25 April 2018.
Haskett, J.A. (2010). Pass-algorithms: a user validation scheme based on knowledge of secret algorithms. Communications of the ACM, 27(8), pp.777-781.
Hebbar, P. (2016). 8 Real Life Examples When Algorithms Turned Rogue, Causing Disastrous Results. Available from https://analyticsindiamag.com/8-real-life-examples-algorithms-turned-rogue-causing-disastrous-results/ Accessed on 25 April 2018.
Hillery, M. (2016) Coherence as a resource in decision problems: The Deutsch-Jozsa algorithm and a variation. Physical Review A, 93(1), p.100-111.
Lee, M.K., Kusbit, D., Metsky, E. and Dabbish, L. (2015). Working with machines: The impact of algorithmic and data-driven management on human workers. In Proceedings of the 33rd Annual ACM Conference on Human Factors in Computing Systems (pp. 1603-1612). ACM.
Lenormand, M., Louail, T., Cantú-Ros, O.G., Picornell, M., Herranz, R., Arias, J.M., Barthelemy, M., San Miguel, M. and Ramasco, J.J. (2015). Influence of sociodemographic characteristics on human mobility. Scientific reports, 5, p.10075.
Luca, M., Kleinberg, J. and Mullainathan, S. (2016). Algorithms need managers, too. Harvard business review, 94(1), p.20.
Medhi, D. and Ramasamy, K. (2017). Network routing: algorithms, protocols, and architectures. Morgan Kaufmann.
Mullainathan, S., Kleinberg, J. and Luca, M. (2017). Algorithms Need Managers, Too. Available from https://hbr.org/2016/01/algorithms-need-managers-too Accessed on 25 April 2018.
Nielsen, T.S., Barros, G.A., Togelius, J. and Nelson, M.J. (2015). General video game evaluation using relative algorithm performance profiles. In European Conference on the Applications of Evolutionary Computation (pp. 369-380). Springer, Cham.
Parikh, N. and Boyd, S (2014). Proximal algorithms. Foundations and Trends® in Optimization, 1(3), pp.127-239.
Pathak, N., Krishnamoorthy, S., Srivastava, J. and Williams, D., University of Minnesota and University of Southern California (USC) (2014). Estimating value of user’s social influence on other users of computer network system. U.S. Patent 8,706,647.
Pringle, R. (2016). When algorithms go bad: Online failures show humans are still needed. Available from https://www.cbc.ca/news/technology/algorithms-facebook-jew-haters-1.4313851Accessed on 25 April 2018.
Schaltegger, S. and Wagner, M. eds. (2017). Managing the business case for sustainability: The integration of social, environmental and economic performance. Routledge.
Ullman, J.D. (2015). Experiments as research validation: Have we gone too far?. Communications of the ACM, 58(9), pp.37-39.
Order an Essay Now & Get These Features For Free:
Turnitin Report
Formatting
Title Page
Citation
Outline
