Problem and Purpose of Model Development - SPRINT Project Model - Case Study Example

MODELLING AND ANALYSIS ASSIGNMENT: SPRINT PROJECT By Student’s Name Code + Course Name Professor’s Name University Cite, State Date Problem and Purpose of Model Development Currently, Hera Comm provides a customer relations manager through numerous ways that involves customer contact desk (CCD), mails, call centres and even websites. Consequently, the organisation has concentrated so much on CCD, which offers a customer with a face-to-face contact with numerous company representatives (Vigo, et al 2014). Hera Comm’s subsequently enjoys about 80 CCDs that have employed about 200 personnel scattered throughout the Northern Italy region serving at least 650,000 customers per year. These CCDs serve a greater level of requests emanating from both the residential and business users by way of providing relevant support requests like the opening and closing of accounts, reviewing and collection of billing facilities and availing information about certain level of services as well as providing resolution on numerous customer complaints (Vigo, et al 2014). It also benefits from a greater size of geographical distribution for the entire CCD network. The desk managers, who supervise each of the CCD representatives, avail relevant and reliable feedback to central planners about the forecasts and proposed schedules as well as provide local situations that might interfere with forecasts. Significantly, they are deemed to be responsible for all operational management as they fine-tune staff schedules in regards to numerous variations (Vigo, et al 2014). This demanding environment has provided an overall problem of ensuring that the customer service is effective and efficient while still making sure to operate at lower costs. Following this line of reason, the organisation has sought to come up with the SPRINT model that would ensure to strike a trade-off between the effectiveness of customer service that is required to attain necessary service levels as well as the efficiency needed to oversee the overall operational cost of the system. It is also important to note that Hera operates in a very competitive market that requires an organisation to formulate certain competitive edges in order to survive the challenges. The need to ensure that it attracts newer customers while still maintaining the loyal 3.2 million customers has ensured that Hera employs a top notch customer relation management (Vigo, et al 2014). Certainly, the SPRINT project arose out of the company’s immediate business requirements. Hera Comm analytical capacities and experience with customer relation management is impressive. The organisation had already CRM systems in place that had allowed it to offer quality services in comparison to other multi-utility companies in Italy (Vigo, et al 2014). The organisation had already a three-level of conducting CCDs so that there was representation in each section of the country. The initial system was composed of CCD representative, desk managers and central planners who had different tasks and objective to meet (Vigo, et al 2014). In 2007, Hera Comm had also engaged another distinctive project, FAST, that was focused on improving the existing quality and costs attributed to operational management especially with the desk managers. FAST model was a preliminary project that provided extensive and significant experience to the organisation as it emphasised on procedural innovations like CCD layout and also, certain other aspects like office automation tools and standardised procedures. In actual fact, the experience the organisation enjoyed from this preliminary project depicted aspects of quality prowess and thus, emphasised on the need for facilitating distinctive innovations by way of implementing on powerful and automated forecasts as well as service optimisation approaches (Bard, Binici & de Silva 2003). It is crucial to note that the Hera had a systematic way of executing CRM services, which perfectly explained on its positive culture towards adoption of rational decision support systems. The culture to employ and rely on rational decision support systems can be seen to have already taken centre-stage in the way CCDs operations were handled. As noted earlier on, the organisation had already established a distinctive geographical distribution of CCDs each reporting to a central placed planner that handled systematic tasks and objectives (Vigo, et al 2014). The value and purpose of the SPRINT modelling to Hera is considered to be two-fold. First, it is expected to offer the already existing central planners immediate access to forecasting and optimisation tools within a user-friendly environment. Secondly, the model is expected to incorporate the different level of processess that are required for the purpose of planning, managing and controlling customer contact desks across the entire network (Vigo, et al 2014). The fundamental stakeholders to the situation at hand include; the desk managers and the central planners. The desk managers are involved in the situation because they are to use the model and ensure that they have partial access to the system components that is needed for operational management purpose (Vigo, et al 2014). On the other hand, the central planners should enjoy full access to the entire components making the system because they are tasked with the responsibility of short and long-term planning as well as have direct responsibility for all operational management functionalities. Hera management provided full and undivided attention to the SPRINT project and termed it as being striving in nature especially in five fundamental key performance indicators like the mean waiting time, PW40, customer satisfaction index, MAPD and, also backlog of back office requests that are made to desk staff (Vigo, et al 2014). The controlling agencies tasked with overseeing quality customer service seems to be in full support of the project and even went ahead to rank the organisation at position 1 for the quality of its CRM services amongst the numerous multi-utility organisations in Italy. SPRINT is deemed to be more of a distinctive collaboration made between the Hera Comm and Optit, which is an academic spinoff firm of the University of Bologna (Vigo, et al 2014). Both Optit and Hera Comm had worked before on numerous projects regarding service optimisation but there is unclear information put forth to expound on whether there were previous projects conducted elsewhere. Assumptions Made In the Development of the Models In developing the SPRINT model, first, it is assumed that the forecasting stipulation must indicate the arrival rate or rather degree of customers at each of the already established customer contact desk over a period of at least one or more months (Vigo, et al 2014). However, it is noted that the mostly-available forecasting process that was formulated for an almost similar workforce management application can only be effective in a short span that ranges between one and two days. For the case at hand, the assumption employed postulates the time span of a single or two months, the total number of staff management assigned to each of the customer contact desk was to be fixed. These CCDs are also expected to remain open from Monday to Friday, which are the normal working days for all multi-utility companies operating in Italy and from 8 AM and 3 PM. The assumption is justified since comparing it to the Bard et al (2003), which proposes of cyclic schedule that are useful for a 7-day-week, 24-hour service, SPRINT model is marred with possible absenteeism on certain days as a result of public holidays or other important functions (Vigo, et al 2014). The model, also, cannot adopt cyclic schedule since the CCDs representatives are employed on different and unique contracts given that some work on full while others work on part-time basis. Secondly, the model operates on the assumption that the working shifts for most of the employees are not longer in comparison to CCDs opening interval and for this reason, they can serve different customer requests (Vigo, et al 2014). These CCDs close at exactly 3 pm and thus, any few customers that are still positioned in the queue for servicing requests are served by any available employees that are re-positioned to additional slots available. The arrangement helps to eliminate the possibility of awarding overtime hours to existing employees (Ernst, Jiang, Krishnamoorth & Sier, 2004). Third, it is assumed that the customer contact desk or rather CCD involves numerous counters and provides customers with a greater level of services that are deemed relevant to fundamental user types, which in this case are both residential and business-based customers. Notwithstanding, it is however; noted that in cases of special situations or even commercial actions the SPRINT model recommends immediate activation of other committed counters so that a single or two additional customer requests can be tackled (Vigo, et al 2014). For instance, it is stated that in the period between 2007 and 2008 Hera Comm embarked on a reorganisation process on their desk services thus developing sets of counters that each served a distinctive customer request. Following this development, the SPRINT model proposed that this reorganisation could be adjusted in order to counter the imbalances caused as a result of under-utilisation of some of the desks. Fourth, the SLA barriers considered are assumed to relate to the mean waiting time of the customers in line as well as the PW40. In fact, the targeted values for the key performance indicators are effectively fixed to necessary values for each of the CCD in relation to the goals and objectives set forth in the SLAs meant for the entire SPRINT project at hand (Vigo, et al 2014). In this regards, possible additional limitations might be set forth on order to prevent the possibility of either opening or closing CCDs for a relatively shorter period of time. It means that a certain counter should remain closed for appropriate level of hours in order to facilitate the CCDs representatives enough time to handle back office duties without any form of interference. Fifth, the SPRINT project model also ensures to develop the service roster for the staff representative positioned within each of the CCDs through a repeated solution approach that takes into consideration a daily optimisation challenge (Vigo, et al 2014). The approach is needed in order to facilitate such aspects as number of counters to be opened, opening profiles as well as the exact number of employees needed to avail appropriate service requests at each of the enabled counter for a certain level of forecasted arrival degrees. Determining the arrivals degrees is justified because it is always highly saturated at any given moment hence cannot allow the model to fully optimise employing conventional techniques like the SIPP. Modelling Content: Strengths & Weaknesses The SPRINT project model has been applied in a clear and cohesive manner. Relevant and reliable data has been established and employed to test its validity. In essence, the input data for the project constitutes detailed information about of the customer that visits the CCDs. The data source has been made available for the period between 2008 and 2009 (Vigo, et al 2014). Another source of data used for the project involves a distinctive group of data that is composed of a number of bills that were each sent to these CCDs at any particular moment in time. The project replaced the historical averages model (HAs), with a hybrid algorithm that employs a M5 model tree in order to predict the daily user arrivals that also incorporates a simple top-down approach (Vigo, et al 2014). The approach has been deployed for the purpose of separating the totals into demand forecasts for specific time allowances and types of customer requests. The use of hybrid algorithm has ensured that the planning results are favourable for the organisation. The organisation has been allowed a chance to effectively forecast each of the arrivals. Markedly, the SPRINT model has helped to significantly reduce the MAPD error objective from a high of 13% to a low of about 12.3% due to system friendliness as well as its capability to develop distinctive yearly predictions for a given CCD (Vigo, et al 2014). The approaches validate the staffing approaches undertaken. For instance, in the development of the adaptive staffing technique to be employed the model predicts user-arrivals peaks by way of a staffing requirement as functionality of the weighted sum of the arrivals in the existing and future time slots. Then again, the notion of the model adopting a single parameter to evaluate the influence of staffing behaviour offers an easier platform to comprehend the approximation methodology. The technique of adaptive staffing allows for an immediate generation of higher quality staffing schedules (Vigo, et al 2014). Furthermore, the simple approximation methodology emphasised, as above, does not prevent the users with little or no mathematical knowledge to establish the perfect compromise level that can be attained between the needed staffing as well as the desire service quality outcomes. However, it is noted that the adaptive staffing that is used in the SPRINT model acts as a fundamental contributory element of successful application of a two-phase optimisation model. With other situations, especially in cases where relatively shorter amount of time is in place, adaptive staffing does not play any major role like in heavily congested occurrences where it limits the SLAs violations (Taylor, 2008). The end results of the study are justified at each of the models employed in the SPRINT project. The justification of these results is validated by way of ensuring the possible alternatives for conducting certain tasks within the systems are eliminated by way of deduction (Vigo, et al 2014). Sufficient information is provided whenever a newer approach is undertaken in order to eliminate possible weaknesses. For instance, it is noted that daily staff-management optimisation process always need significant level of resources that are focused on the service offered to customers arriving within a given day of operation while at the same ensuring to uphold the effected target values for all SLA-related KPIs as well as any other possible operational constraints. Modelling Process The modelling process of the staffing-scheduling optimiser employs a fast simulator of the SE module since adaptive staffing mechanism will solely implicate approaches associated with beta. The adoption of the project involved a greater number of stakeholders that included the Hera Comm management team, the researchers from Optit, desk managers, the central planners as well as the CCDs representatives (Vigo, et al 2014). All of the people had distinctive prior experience in operating contact desks and even had a formal way of executing tasks from a central location managed by central planners. The project design employed has indeed used a significant number of other useful models that relate to each other in providing quality plans and accurate demand forecasts (Vigo, et al 2014). In essence, the design has allowed the elimination of unnecessary costs like overtime pay by using the existing personnel to undertake both front and back office tasks within certain well laid-out timeframes. The most notable constraint that can be perceived within this model system rests with complexity of finding out the parameters needed to expound on the demand forecasting process. In fact, the process of establishing the objective function is deemed tedious (Vigo, et al 2014). Other parameters like M have been inappropriately used within the model to represent certain features of the demand forecasting helps resulting to informational gap on the same. Both the project and client’s expectations have been effectively managed throughout the modelling process. In fact, the clients have enjoyed a great number of benefits that include; efficient exploitation of management data that were previously employed in formulating management summary reports, there has been increase sales and reduced costs while central planners have now been allowed a single planning strategy that is easily manageable. References List Bard J, Binici C & de Silva AH 2003 Staff scheduling at the United States Postal Service. Computer Operational Research, Vol.30, no.5: pp.745–771. Ernst AT, Jiang H, Krishnamoorthy M & Sier D 2004 Staff scheduling and rostering: A review of applications, methods and models. European Journal of Operational Research, vol.153, no.1:pp. 3–27. Taylor JW 2008 A comparison of univariate time series methods for forecasting intraday arrivals at a call center. Management Science. vol.54, vol.2: pp. 253–265. Vigo, D et al 2014 SPRINT: Optimization of Staff Management for Desk Customer Relations Services at Hera. Interfaces vol.44, no.5: pp.461-479 Read More