# Introduction roductivity refers to the physical relation between the quality produced (output) and the quantity of resource used in the course of production (input) (Susan, 2009): Productivity (P) = output (O)/ input I Output implies production while input means land, labour, capital, management etc. Productivity measures the efficiency of the production system. Higher productivity means producing more from a given amount of input or producing a given amount with minimum level of inputs. In other words the more the output from one worker or one machine (or a piece of equipment) per day per shift, the higher is the productivity (Susan, 2009). Higher productivity is not to be taken in sense of higher workloads or faster machines alone but it is always elimination of waste of all type of labour (time and skill) machine time, capital, and material management etc. # Productivity = Output per unit of input Productivity and production are two different terms. Productivity is a relative term indicating the ratio between total output and the total inputs used therein on the other hand production is an absolute concept, which refers to the volume of output (Gerard and Bart, 2009). The volume of production may increase but productivity may decline due to inefficient use of resource. Efficient use of input may increase productivity but the volume of production may not increase. Production refers to the end result of production system where as productivity reflects its efficiency. The benefits of productivity includes: It helps to cut down cost per unit and thereby improve the profits, gains from productivity can be transferred to the consumers in from of lower priced products or better quality products, productive entrepreneur can have better chances to exploit export opportunities and generate more employment opportunities (Gerard and Bart, 2009). Productivity may be measured either on aggregate bases or on individual basis, which are called total and partial productivity respectively. # Total Productivity Index= total outputs/ total inputs This index measures the efficiency in the use of all the resources. Partial productivity Indices, depending upon factors used, it measures the efficacy of individual factor of production (Andersson, 1996). Workers' productivity can be tied to these parameters (OECD, 2002): # a) In-time Completion of task The completion and delivery of a particular task from a hand of a specific employ is tied to the overall success of the organization. It determines if the organization project will be completed in time or not. Intime completion of task is an integral criterion for determining the productivity of a particular employee. # b) Duty Punctuality Punctuality is tied to organization success. How can an organization attain successful, when most employees are not punctual? A particular task not completed in-time result in total delay of the overall organizational project which is highly expensive. # c) Optimal Interaction with staff Staff cannot, and will not work in vacuum or in isolation. The specification requirement of a small subsystem must be aligned with the overall system project. Therefore discords among staffs, hamper project success and hampers productivity. A particular employee cannot be productivity if he cannot work, closely with other staff. # d) High Maintenance Culture Organizational tools and equipment are meeting for productive organizational duties. Breakdown of tools and equipment by employees will result less productive activities. Therefore if an employee cannot maintain highly tools and equipment in an organization he or she is not productivity. # e) Minima Dissension with staff Tolerant is very vital in any organization for it to succeed. Senior executive must tolerant lower level staff and lower level staff must tolerant senior executive failings leading to minima dissension and promote productivity. # f) Improved Technical, Business and people Knowledge Constant training, attending of seminar, workshop and conference will improved the knowledgebase of any employee, which if applied collectively will improve productiveness of an organization. # g) Extra Secular Activities All work with no extra secular activities, like after work get together will create depression, disorder and intolerant among staff. Extra secular activities must be encouraged. This research paper is geared toward proposing implementing worker productivity: a fuzzy supervised neural training algorithm approach. # II. # Review of Related Literature The theory of fuzzy logic provides a mathematical strength to capture the uncertainties associated with human cognitive processes, such as thinking and reasoning. In standard set theory, an object does or does not belong to a set. There is no middle ground. In such bivalent systems, an object cannot belong to both its set and its compliment set or to neither of them. This principle preserves the structure of the logic and avoids the contradiction of object that both is and is not a thing at the same time (Zadeh, 1965). However, fuzzy logic is highly abstract and employs heuristic (experiment) requiring human experts to discover rules about data relationship (Angel and Rocio, 2011). Fuzzy classification assumes the boundary between two neighboring classes as a continuous, overlapping area within which an object has partial membership in each class (Kuang et al., 2011). Fuzzy logic highlights the significant of most applications in which categories have fuzzy boundaries, but also provides a simple representation of the potentially complex partition of the feature space. Artificial Neural Networks (ANNs) constitute a class of flexible nonlinear models designed to mimic biological neural systems. An ANN is a mathematical model or computational model based on biological neural networks (Gutiérrez, 2011), as an interconnected group of artificial neurons, which carries out computation using a connectionist approach. Typically, a biological neural system consists of several layers, each with a large number of neural units (neurons) that can process the information in a parallel manner. The models with these features are known as ANN models (Robert, 2000). ANNs have been widely applied to solve many difficult problems in different areas, including pattern recognition (matching), signal processing, language learning, electronic medical record processsing, tele-diagnosis and computer networking (Robert, 2000). Neural network utilize dataset. The data set is divided into three distinct sets: training, testing and validation sets. The training set is the largest set and is used by neural network to learn patterns present in the data. The testing set is used to evaluate the generalization ability of a supposedly trained network. A final check on the performance of the trained network is made using validation set. Learning methods in neural networks can be broadly classified into three basic types Supervised, unsupervised and reinforced learning (Diogo et al. 2008). Supervised learning is the machine learning task of inferring a function from supervised training data. The training data consist of a set of training examples. In supervised learning, each example is a pair consisting of an input object (typically a vector) and a desired output value (also called the supervisory signal). A supervised learning algorithm analyzes the training data and produces an inferred function, which is called a classifier (if the output is discrete) or a regression function (if the output is continuous). Unsupervised learning studies how systems can learn to represent particular input patterns in a way that reflects the statistical structure of the overall collection of input patterns. By contrast with Supervised Learning or Reinforcement Learning, there are no explicit target outputs or environmental evaluations associated with each input; rather the unsupervised learner brings to bear prior biases as to what aspects of the structure of the input should be captured in the output. Unsupervised learning is important since it is likely to be much more common in the brain than supervised learning (Benedetti et al., 2005). Reinforcement learning, one of the most active research areas in artificial intelligence, is a computational approach to learning whereby an agent tries to maximize the total amount of reward it receives when interacting with a complex, uncertain environment. In Reinforcement Learning, provide a clear and simple account of the key ideas and algorithms of reinforcement learning. Their discussion ranges from the The two most widely used neural networks are the feed-forward networks and recurrent or interactive (feedback) networks, kohonen's self-organizing network, Adaptive resonance Theory (ART) and Counter propagation network are others (Chakraborty, 2010). Feed-forward ANNs allow signals to travel one way only; from input to output. There is no feedback (loops) i.e. the output of any layer does not affect that same layer. They are extensively used in pattern recognition (Chakra borty, 2010). This multi-layered structure of a feed-forward network is designed to function as a biological neural system. The input units are the neurons that receive the information (stimuli) from the outside environment and pass them to the neurons in a middle layer (i.e., hidden units). These neurons then transform the input signals to generate neural signals and forward them to the neurons in the output layer. The output neurons in turn generate signals that determine the action to be taken. It is important to note that all information from the units in one layer is processed simultaneously, rather than sequentially, by the units in an "upper" layer (kuan and white, 1994). # a) Feedback Network or Recurrent Neural Networks Feedback networks can have signals travelling in both directions by introducing loops in the network. Feedback networks are dynamic; their 'state' is changing continuously until they reach an equilibrium point. They remain at the equilibrium point until the input changes and a new equilibrium needs to be found (Chakraborty, 2010). Kohonen's Self-Organizing Network is a twolayer, feed-forward network (Beale and Jackson, 1990 and Dayh off, 1990).The first is an input layer and the second is a grid or map arranged in a one or twodimensional array. The second layer is known as a competitive layer. Incoming patterns are classified by the nodes that they activate in the competitive layer. Similarities among patterns are mapped into closeness relationships on the competitive layer. After training, the pattern relationships and groupings are observed from this layer. Adaptive Resonance Theory (ART) is an unsupervised, competitive learning algorithm (Beale and Jackson, 1990).It is a two-layer network arranged in feedback and feed-forward connection. The layers have different functions, unlike the Multilayer or Kohonen networks. The first layer can be either an input or a comparison layer and the second layer can be either an output or a recognition layer. Both are interchangeable during training. # III. # Methodology and Design Existing approaches in determining worker productive are based on classical set method which usually tied precision to these variables. This is usually flawed in approach because the parameters for accessing worker productivity are imprecise; therefore fuzzy logic will handled this approach very well. Numerous algorithm has be proposed for solving real worker productivity problems such telecomputing through telematics, but still date few Fuzzyneural network algorithm has be proposed for objective recognizing worker productivity. IV. # The Proposed Fuzzy Supervised Neural Network Training Algorithm Approach The proposed Algorithm imbibes artificial intelligence techniques in tying the parameters for identifying worker productivity into a learning paradigm thereby establishing a conclusive boundary. Unlike the current approaches, in which success or failure are based on the wills and experiences of relevant personnel designing and administrating the approach in other to elicit relevant recognition points, success and failure in this approach are not dependent on human intuitions, but success, is closely linked within tuned-up approaches within the carefully and systematic implemented algorithm variables. The Algorithm is depicted on Figure 1 Worker productivity: Target Result ("Productive Worker", "Might be Productive Worker" and "Not Productive Worker") # Implementation and Discussion The implementation of our result was dual fold; the neural training dataset was handled conveniently utilizing Matrix Laboratory (MATLAB) which serves as our simulation tool in achieving the our results because of its interactive environment for algorithm development, data visualization, data analysis, and numerical approach which was relevant to our numerical dataset which was more appropriate than with spreadsheets or traditional programming languages, such as C/C++ or Java. After pruning the dataset utilizing MATLAB, the algorithm was fully implemented utilizing Hypertext Preprocessor (PHP), which served as the language of implementation. # VI. # Discussion The implemented algorithm provides an interactive base in determining varied worker productivity objectively as opposed to the subjective approach which is achievable utilizing otherapproaches. The result was satisfactory having been able to distinctly determine worker productivity. # VII. # Conclusions This paper has demonstrates the practical application of fuzzy supervised training algorithm for worker productivityin various organization. # Global ![(Sun andJang, 1993 andAhmad, 2011) Conventional approaches of pattern classification involve clustering training samples and associating clusters to given categories. The complexity and limitations of previous mechanisms are largely due to the lack of an effective way of defining the boundaries among clusters. This problem becomes more intractable when the number of features used for classification increases (Christos and Dimitros, 2008).](image-2.png "") // Initialization WP (P0)1. Randomly pick a Worker K;2. Save Target Result in Knot;// Loop till terminal point3. While WP (P)<>7do;// Not Productive Worker1. If WP (P1); in-time completion of task is high, while other P is Low or exempted THEN NotProductive Worker2. If WP (P2); in-time completion of task and duty punctuality is high, while other P is Low orexempted THEN Not Productive Worker3. If WP (P3); in-time completion of task, duty punctuality and optimal interaction with staff is high, while other P is Low or exempted THEN Not Productive Worker 4. Else 5. End IfYear 2014// Might be Productive Worker6. If WP (P4); in-time completion of task, duty punctuality, optimal interaction with staff andhigh maintenance culture is high, while other P is Low or exempted THENMight beProductive Worker7. Else8. End If// Productive Worker9. If WP (THEN Productive WorkerG12. Else13. End If14. End © 2014 Global Journals Inc. (US) Worker Productivity: A Fuzzy Supervised Neural Training Algorithm Approach * Fuzzy Approach to Likert Spectrum in Classified levels in Surveying Researches HAhmad 2011 * Employee Cynicism: An Examination Using a Contract Violation Framework LAndersson Human Relations 1996 * CAngel RRocio Documentation Management with Ant Colony Optimization Meta 2011 * Introduction to Neuro-Fuzzy Systems FRobert Advances in Soft Computing Serie Berlin/Heidelberg, Germany Springer-Verlag 2000 * A Neuro-Fuzzy Classifier and its applications CTSun JSJang Proc. IEEE Int. Conference on Neural Networks IEEE Int. Conference on Neural NetworksSan Francisco 1993 * International Comparisons Of Hours Worked: An Assessment Of The Statistics FSusan Monthly Labor Review 2009 * LAZadeh Fuzzy Sets. Information and Control 8 1965