To return to previous page, click on: Techniques.htm
0) Project Planning Tools: These tools include the Gantt chart, the PERT chart, the Work Breakdown Structure chart based on your expectation of what tasks should be done, when they should be done, who will do them, methods to use, and how much time it should take to do each task and for each person in the group to spend on the project.
1) Process Flow Chart: This chart is to be done as early as possible to understand the flows of products, services and information through the organization.
2) Historical Chart of key events: create this chart from the interviews conducted with workers and managers; this could provide the researcher with understanding of what caused what problems in the organization.
3) Problem Definition: Not a technique but an important step in determining the actual situation, the goal state, and the gap which determines the problem to be solved. If there is no gap, there is no problem.
3.1) Fishbone Diagram: a cause-effect analysis in diagrammatic form; can result from the historical chart, process flow chart, interviews, and observation; feeds into the problem definition.
4) Data Collection and Organization: Data can be both qualitative and quantitative; cross-sectional or longitudinal; obtained from files, records, observation of events, interviews and published documents.
4.1) Data Organization Checksheet: Very important tool to organize your information in tabular form for subsequent statistical analysis; should be used before analysis begins. You should answer the questions: what data do I need; what does the organization allow me to collect; what surveys should be carried out; open-ended or closed-ended or both; will you do a sample of the population or the entire population; how will you sample; if past data, how many days, weeks, months, years in the past will you record in your checksheet; what will be the rows and columns of the checksheet; how many checksheets will you use; will you compare the organization to itself in previous time periods or with other organizations at the same time period; how will you massage your data; will you use numbers (dollars or customers or sales units) or ratios of numbers (dollars per unit); sources of data which you are comparing to your organization (industry data versus company x data).
5) Analyzing the Data: There are a number of tools that can be used to analyze the data on the checksheet.
5.1) Pareto Chart: A histogram which can identify the entities that are causing the largest percentage of problems; Pareto’s Principle states 80% of the problems are caused by 20% of the entities; these are the entities that most likely are producing the bottlenecks.
5.2) Run Chart: A plot of y versus x over a period of time; y is the variable indicating throughput (customers served, products sold) , defects or some other performance variable; x is time in the appropriate period (days, weeks, months, quarters, years -- depends on your problem and the available data); this chart tells you if there is a pattern in the performance over the time period plotted; patterns help you analyze the underlying causes of the problem.
5.3) Histogram: Vertical bar chart that is a very general tool that is used to plot the values of system behavior or performance or characteristics versus an independent variable on the x axis (entity on the production line, branch of the company, time period, etc.)
5.4) Scatter Diagram: This is a method to identify whether there is a relationship between the y and the x variables when data are plotted as points on a graph. Suppose you desire to explore whether one variable (sales, costs, complaints, defects, units produced, customers serviced, papers processed) depends on some other variable ( calls received, advertising expenditures, money spent or time period). If the relationship is postive, then the best straight line through the data points will be upward sloping and vice versa. From the scatter diagram, you can obtain the correlation coefficient (r ) ,the coefficient of determination (r squared) , the equation relating the y to the x, the ability to forecast the y variable if the x variable is extended.
5.4.1) Forecasting Techniques: A variety of forecasting techniques enable the researcher to produce an amalgamated forecast that combines the results of different methods according to how well they forecast the recent known past.
5.5) Control Charts: used to plot a variable from a manufacturing process against time to see if the variable stays within a certain bandwidth around a mean target level. The variable could be the number of defects, the size of a part, the desired number of units produced by a station, etc. The bandwidth is determined by the confidence level which you set. The bandwidth is based usually on the assumption of a normal distribution of the variable at any point in time.Using the normal distribution and the desired level of confidence, the upper and the lower limits of the variable can be set. When the variable exceeds the bandwidth, it is called out of control which indicates corrective action should be taken.
6.) Determining the solution to the problem: Once the analysis is completed, the researcher will need to recommend a solution to the problem. This is done using a number of tools and techniques.
6.1) SMART Chart: Used to quantify the qualitative; used to decide which alternative solution is best in solving a problem dealing with quantities and qualities (numbers and non-numerical information); forces researcher to think of alternative solutions and to think of the criteria for judging a solution. Both the selection of the criteria and the generation of the alternative solutions must be done by human intelligence. In addition, the estimates of the weights and ratings in the SMART chart require human judgment, albeit justified by a narrative argument. Weights could also be estimated using a Confidence Chart, as explained in the folllowing paragraph. The calculation of the best solution can be done by obtaining a score for each alternative and choosing the alternative with the highest score.
6.2) The Confidence (or Attribute) Chart: Attributes are the criteria used to evaluate the alteratives in the SMART chart. This chart requires the researcher to compare each pair of attributes to determine the relative importance of one attribute with respect to the other. This is done for all pairs; the results are place in the cells of the chart. Points are assigned to each cell based on the relative importance symbol in the cell. Along each row the points are added up and placed in a raw weight column. Raw weights are normalized to 100 in the right-most column. These weights are used in the SMART chart.
6.3) Poka Yoke: The solution to an organizational problem requires identification of the root causes of a problem and then a way to eliminate the root causes. One method is to determine why errors occur and to design the production system so that it is highly unlikely or impossible for the error to occur . To do this, the designer must imagine an operator or a customer thinks when producing or using the product or service incorrectly; then he must design the system such that the operator or customer will not be able to make the error.
6.4) TOC (Theory of Constraints): This is the theory of identifying and solving bottleneck problems in the production process. It involves observing the process as it takes place on the factory floor or the service center; taking data on the time to produce a product or service; linking accounting information with manufacturing information with sales information; with observing information feedback loops; and noting where inventory is piling up or customers are waiting to be served. By installing “ropes, quality control points, drums and buffers,” solutions are designed to solve the bottleneck problem.
6.5) Present Value Analysis: a financial tool used to analyze financial data of costs and revenues for alternative solutions to a problem. By bringing all financial flows in the future back to one point in time, the present point, the alternatives can be evaluated and a choice made.
6.6) Decision Trees: A method of combining the probabilities and the payoffs of outcomes in a decision situation where several possible events could occur as a result of different actions of the manager. It involves a combination of hard data and judgementally assessed probabilities to arrive at the solution with the highest expected value.
7.) Implementation: A Political Implementation Flow Chart can be used to indicate how the chosen solution for the problem can be implemented within the organization.
8) Lessons Learned: Not a technique but a method of introspection in which the researcher analyzes mistakes or false starts, wasted time, blind alleys, red herrings and explains what he would do to avoid these mistakes if he could advise a student in a similar project; also should include which choices the researcher had made correctly.