Anova in Supply Chain and Logistics Management - Case Study Example

Anova in Supply Chain and Logistics Management Name: Institution: ANOVA in Supply Chain and Logistics Management 1(i) Quantitative Concepts and methods covered in OMGT 2816 The present business environment can be characterised by fast developing ICT and general speed of changes. These rapid changes are bound to have a significant effect on supply chain management and logistics. Quantitative methods can play a pertinent role in meeting high demand and optimizing a company’s resources. Various quantitative methods can be used in the supply chain to satisfy the customer, which is the main aim of any supply chain and logistics system. These models methods involve models that can be used in controlling and replenishing of inventories with the aim of determining order quantities to minimise the total average costs. Four main methods have been analysed in the studied unit. Simple Regression analysis- this examines the strength and relationship between two variables using the correlation coefficient and scatter plots. By formulation of hypotheses, the slope of the model can be analysed. Multiple Regression Analysis- this examines the strength and relationship between one dependent variable, and many other independent variables. This method involves applying significance tests on the regression model and its coefficients. These significance tests can evaluate which variables contribute most to the model and which have little or no influence on the model. This method also involves computing and interpreting the residuals, the coefficient of determination and the standard error of the estimate. ANOVA experiment design- this is a test for overall experimental effect. It can be used to run a comparison of means between multiple groups at one time. It is mostly used to determine if whether the population means of the dependent variable for several sample are the same or if there are significant differences between two or more population means. Time series analysis- this involves the use of forecasting models to generate trends for different types of forecast. Under this method, there are three types of components for a time series. Trend component which is the long-run general direction of the value of a given variable over a certain period of time; cyclical component which involves rise and fall in time series data that cover long periods of time; seasonal component which observes the behaviour patterns of data that occur In periods of one year or less. Logistics is the process of strategically managing procurement, movement and storage of already manufactured goods; this is achieved through organisation and marketing channels. This process also puts into consideration cost effectiveness and profit maximisation of company profits. This is essentially a framework that creates a design for the flow of information and products through a business. Supply chain management builds upon this logistics framework and serves to link the suppliers and the customer. The main aim of supply chain management is to safeguard relationships in order to have a profitable outcome for all participants in the chain[Chr13]. For making the right and effective decisions in logistics, functional operations and proper systems are required. Decision making in the supply chain is the core of any organisational or management system. Decision making has become very complex in the 21st century due to the rapid growth of technology. Therefore, decision making based on proper business models must be very important in the field of logistics and supply management. In logistics, decision problems are different from any other functional departments in an organisation. This is mainly because logistics are carried out between the organisation and its different supply chain partners. The need to communicate and collaborate between logistics operators in the same supply chain is what makes the decision-making process such an intricate procedure. This is why proper quantitative methods need to be incorporated to ensure the right customer receives the appropriate quality and quantity at the right place at the right time with minimum costs. 1 (ii) Analysis and problem solving in supply chain and logistics management A distinctive supply chain and logistic problem involves a manufacturer who gets components from different suppliers. These components are then sent to a distribution centre where they are finally assembled and transported to the areas where they are demanded. Supply chains have been regarded as complex systems; a high number of factors affect this complex system. The ways these factors interact over time increase the complexity of many supply chains in different companies. Therefore, modelling has been recognised as the preferred method in solving related problems that may affect these complex systems[Lon11]. In solving these problems companies can effectively diagnose problems, train personnel and managers, explore possibilities, choose correctly and transfer research and development results to real functional systems. Another problem that adds to the main problem of complexity of supply chains is the one that involves using the correct supply chain to the preferred product, some supply chains may be expensive to transport products to the market and this is a troubling issue in the supply chain and logistics sector. Another angle the decision-making process can be looked at is in supply chain, and logistics management is the marketing perspective. The result of any supply chain is the market or the customer. Analyzing different marketing strategies may involve asking if different marketing strategies are affected by the size or activity of the supply chain. Also, questions as to which type of supply chain may be used, or a combination of which would be preferable. Another question that may frequently arise is can two supply chains be used to deliver a single product. 1 (iii) Literature review The ANOVA method for variance analysis has proven to be one of the most effective tools in supply chain management. This can be attributed to its ability to compare different sample means and from set hypotheses, conclusive decisions can be made. A classic example of the use of ANOVA is where average profit margins and costs can be compared between different products to find which has the highest profit with respect to the production cost. When investigating various factors on the quality of a supply chain, [Mad08] Used ANOVA analysis on a supply chain model that was examining the demand uncertainty, distribution issues, supply chain operations and supply chain speed. In their study, the effects of crucial supply chain and logistics factors on quality and speed of a supply chain network were investigated. In their documented journal, the authors argued that for a company to gain competitive advantage, managers need effective supply networks to understand the challenges of speed and quality in the modern error. Although supply chain quality management appears to be one of the major determinants of supply chain effectiveness, not much research has been put on the crucial success factors and their impact on important chain performance indicators[Mad08]. During the first stage of the conducted experiment, group factors that were found to be significant were broken down into single factors in the follow up experiments. The critical issues in the sphere of their simulation were defective rate reduction, supplier involvement and supply chain speed. The findings of their research showed that to achieve supply chain quality, continuous improvement needs to be adopted. Second, it was also noted that supply chain competencies played a significant role in the supply chain setting. The overall results showed that demand and supplies uncertainties did not determine the viability of the supply chain networks. Instead, the viability of supply chain networks is mainly determined by supply chain quality[Mad08]. In analysing the relationship between business intelligence and supply chain management for strategic marketing decisions,[Ser14] used ANOVA analysis to compare differences in mean values of variables in relation to the size, legal form and activity. After finding a significant statistical correlation between business intelligence, information visibility, supply chain management and integration among production chain partners, the analysis was conducted. The analysis showed that statistically significant differences in the mean of business intelligence were obtained only when analyzing the activity of the company[Ser14]. Various supply chains might be mismatched with various products, [Nag10] Carried out research to find out if there was a mismatch between the product and the supply chain used. She also carries out research to establish if there are innovative and functional products that match effective supply chains that respond to the market. Nagy refers to a volatile market as one which customer expectations change in a dynamic and unpredictable manner; this can be the reason companies may mismatch a product and a supply chain. By using ANOVA, she was able to establish that there was not a significance difference between the groups under investigation. Her next analysis involved the mismatch reason of companies aiming to attain better performance by combining the strength of two supply chains. The ANOVA analysis showed that combining to good supply chains cannot be accepted as a reason for the mismatch. [Nag10] In her research, carried out cluster analysis to separate companies into those that apply demand chain tools at a superior level and those that apply them at an inferior level. The next stage involved describing the characteristics found in the two clusters. She then used ANOVA mean comparison on a 95% significance level. The results showed a significant difference in the case of almost all demand chain management tools used. This meant she could now come up with two clusters, one with companies that use superior demand chain management tools and one with inferior demand chain management tools. The results of further analysis showed that companies with superior demand chain management tools consisted of mainly large companies. In the category of tools such as performance and management costs, customer assessment was significantly different between the two groups. Results also showed little emphasis was laid on information management tools by companies with inferior demand management tools. One of the major concerns of supply chain management today is the study of inventory systems. As soon as the parameters affecting the supply chain are identified, simulation plays an important role in evaluating the best the best strategy involving the variables lead time, inventory policies, demand patterns, transportation costs and customer satisfaction. Modelling using techniques such as ANOVA become very crucial in such scenarios[Lon11]. In his research, Longo used ANOVA analysis to analyze nine stores, three plants, four distribution centres and twenty items. After finding the significant variables that affected the 9 stores, the results from the analysis enabled correct arrangement of the warehouse internal resources in order to maximize the average number of packages handled in a day and to minimise the total logistics costs[Lon11]. The inventory problem is not solely not responsible for the problems caused to supply chain networks; internal logistics management at every stationed supply node also affects the entire supply chain system[Lon11]. [Rao11] Tried to investigate important determinants for information technology adoption and test the value of supply chain organisation and co-ordination strategies in improving performance. In their analysis, three hypotheses were used to test the relationship between IT adoption by firms, information coordination with suppliers, supply chain co-ordination and information coordination with the market or the customers. IT adoption was regarded as the dependent variable while coordination with suppliers, supply chain coordination and information coordination with customers were regarded as independent variables. One way ANOVA tests were carried out to test if each strategy affects the entire performance. The results showed that supply chain coordination, coordination with customers and coordination with suppliers were very successful in explaining adoption of IT. The ANOVA analysis showed that all this aspects reflected a positive relationship with organisational performance[Rao11]. 2. (i&ii) Supply chain and logistics management problems and the application of ANOVA to these problems One of the major problems that require appropriate decision making in supply chain management is the chain inventory problem. At any chain supply node, any inventory management system has to answer three questions that are when to order new products, what quantities of this products and how often to review the stock status. Supply chain managers have to take into account the notion of an extended enterprise. Extensive research has to be adopted to consider the inventory management problem. The analytical methods that have been constantly adopted have not been able to accommodate all the dynamic supply chain variables that keep changing and become more complex to analyse. The study of inventory systems in supply chains is one of the major concerns in present day supply chain management. When the number of parameters affecting the supply chains performance becomes high and focus is placed on the emphasis of the supply chain analysis, simulation may become a key aspect in finding the proper balance between transportation cost, demand patterns, inventory policies, lead times and customer satisfaction. Here simulation combined with statistical practices like the use of ANOVA can be used to analyse such supply chains. Since ANOVA is mostly used in detecting if there is a significant difference in two or more populations, in such a scenario, it can be used to understand how the parameters affect the supply chain behaviour. Statistical instruments such as ANOVA can be very useful in giving support to simulation models. 2 (iii) ANOVA applications in the inventory problem The parameters or factors that mainly affect a supply chain are lead times, transportation cost, demand patterns and inventory policies. Using ANOVA can test how all this factors affect the supply chain network. It can also be used to understand the impact of these parameters on performance measures. Managers can assess the importance of certain factors on the supply chain and make correct decisions on which parameters should be considered when faced with a problem. 3 (i) Limitations to the ANOVA method While this method may be useful in comparing products and processes that would be most convenient and those that would reduce costs and maximise profits, it is limited to some extent. In order to use the ANOVA analysis method, some assumptions have to be made. First, the data under inspection has to be from interval or ratio scale, in other words, the data has to constitute of continuous variables. Data from interval and ratio scales have the characteristics of nominal and ordinal data but are different in that interval data is predetermined by an interval and ratio data is not based on a true zero point. Data that is classified into purely nominal or ordinal scales may prove taxing to examine by use of the ANOVA method. Second, data that is analysed using the ANOVA analysis is mainly assumed to be normally distributed, this means that the data is meant to follow a normal distribution. This is not always the case as data may tend to follow other distribution and transformation of this data into normal distributions may prove to be difficult depending on the distribution of the data. The third assumption is that there is independence in the errors if this assumption does not hold the ANOVA model will not be executed. While comparing some subjects using the ANOVA analysis, many factors may affect the data. For instance, if, in the logistics field, one may want to compare the efficiency of males as compared to females at a certain supply chain and logistics node. Respondents may be subjected to different factors during a particular time of the year, and this may give inconclusive results. For instance, participants may improve their work output within the time, and this could spoil earlier research at some particular time that could have been already implemented. For instance, if a particular group is found to have better output at a particular time as compared to another, their performance may change due to factors that the supply chain manager is not in control of, e.g. they may be more motivated to meet deadlines due to personal reasons. Participants may also become bored and decide not to continue with the same efficiency; this has been seen in plants or organisations where the work is too monotonous or an organisation that experiences a high labour turn- over. Employees employed in such a field mostly lose confidence and motivation since they are uncertain of the future. Since ANOVA involves comparing the mean certain groups, a collected score can be used to show the influence of a certain factor over different levels. By using the ANOVA table, a single score may have a positive effect on a certain level A and also have a positive effect on a level B. the same score may have a negative effect on a level C making it difficult to analyse data. It is also not always possible to compare using the ANOVA table, for example, while it may be possible to compare men and women in a certain supply chain and logistic scenario, it is impossible to compare them in some areas such as managerial levels where the results may always prove not to be significantly different. 3 (ii) solutions to the limitations offered by the ANOVA method The main limitations that are assigned to the use of the ANOVA method are mostly finding data that can relate to the assumptions of the ANOVA table. Though using this method may seem difficult to use, it has been simplified by the use of computer statistical packages that make it easier to make deductions. Data manipulation is one key aspect that may be used to make sure that the data is free of any error and is normally distributed. Through mathematical procedures, data that is not normally distributed can be transformed to be normal. This may call for supply chain and logistics departments to improve their IT resources so as to cope up with better ideas to solve problems. Models have not been used mostly in the recent past, but now supply chain departments are adopting these methods to reduce the complexity of the supply chain process[Lon11]. Knowledge of required statistical package may come in very handy to analyse complex supply chains. Managers should also determine the best groups for comparison as some situations may not be ideal, for example comparing men and women may not give proper results as they both may perform the same duty well in the same conditions. References Chr13: , (Christopher, 2013), Lon11: , (Longo, 2011), Mad08: , (Madu, Kuei, & Wich, 2008), Mad08: , (Madu, Kuei, & Wich, 2008), Ser14: , (Seric, Rzga, & Luetic, 2014), Nag10: , (Nagy, 2010), Rao11: , (Rao, Sahu, & Mohan, 2011), Rao11: , (Rao, Sahu, & Mohan, 2011), Read More