Tmgt 361assignment Vi Instructionslectureessayrandomnesswhat Does Ran
TMGT 361 Assignment VI Instructions Lecture/Essay Randomness What does random mean? It means it wasn’t controlled, that mere chance cause it (whatever it is). Random is not the same as merely being mixed, or disorganized, or no discernable pattern. If I take different densities of different sizes of pebbles and I put those pebbles in a bottle and I gently shake the bottle, the pebbles, due to their size and shape and density and how I shake the bottle will settle. Though I might be able to discern any pattern, though a different mix of pebbles or a different type of shaking could result in a near infinite settlings, the settling wasn’t random.
If I use all the same size and shape and density of pebbles, and I shake the bottle enough that every pebble has an equal chance of being in every position—now I achieve, or for all practical purposes, achieve a random arraignment of pebbles. Randomness is not achieved unless every possible outcome (every pebble location, every coin or die toss outcome, every card that can be dealt) that can occur has an equal change of occurring. If any outcome has a greater chance of occurring than another outcome (e.g., heads more than tails on a coin toss) we know instinctively that something is not right. The reason a coin toss is considered fair is because it is random, i.e., there is no impartiality or bias in the event, heads and tails had an equal and random probability of occurring.
There is an argument that noting is truly random, that if we enough information we could predict everything perfectly. Philosophically this may be true but we live in a real world where multiple forces and sequences of events affect things to the point where we can’t predict coin tosses (using fair coins and tossing the coins in a random fashion). Numerous, mostly undetectable and/or uncontrollable forces and imperfections influence every process and outcome. We cannot make a perfect circle nor a perfect drill bit. We cannot perfectly repeat every shot of the basketball.
We cannot make two perfectly identical nails or ballot moves. Do not think that every mistake or error is random; most are not; they have a cause, a special , assignable , or attributable cause (even though we may be unaware of the cause). We cannot do anything about random error (which may be termed common cause or expected error). The oldest and core part of quality as a profession or discipline is to reduce non-random error. Therefore, it would be very useful to separate random errors (which you cannot control or eliminate) from non-random errors.
Sometimes the cause of the error leads to an obvious and clear mistake. Often, mistakes are not clear.
Often we don’t know something is a mistake. Think of the seemingly random nature of people getting sick before we knew there were germs and how to pasteurize. There was a cause and we didn’t know it.
Similarly, most actions, products, and processes are affected by random and non-random error. Statistics have greatly helped us separate special cause from common cause error. With statistics, we can be certain of the probability that something is random or not—we may not know or every know the cause of the non-random error or how to prevent it, but we can still identify it. We will learn about statistical tests in a future lecture. Many tables, charts, and graphs that we use to make quality decisions are based on statistics and the probabilities of outcomes.
Without, the table (or the math calculations that created the table) we wouldn’t know if 2 heads in row is unusual enough that we label it as non-random. Two heads in a row happens 25% of the time. If we labeled this as non-random, we would have made a bad conclusion most of time. Five heads in a row only happens due to randomness 3.125% of the time. Are we willing to label a 3% random occurrence as non-random.
If I flip a coin 5 times and get 5 heads are you going to accuse me of cheating? If one hundred people each toss 5 coins, the probability is that 3 of those people will toss 5 heads. Do you accuse those three of cheating? Ten heads in a row naturally, randomly occurs slightly less than 1 time in a 1,000. If the card or lottery player makes a 1 in a 1,000 or 1 in a billion win, was it cheating?
Maybe. What we would do determine (usually statistically, based on probabilities) if the outcome was unusual and then see if the cause can be determined to be random or assignable. Sometimes we just go with the odds. If I am monitoring a manufacturing process and something happens (goes wrong) that only randomly happens 1 in a 1,000 times, the other 999 times it happens it will not be random. If I take action (which takes time and money and slows production) to fix thing, I will only be trying to fix something that cannot be fixed (for random error cannot be changed) 1 in a 1,000 times.
Note the decisions have to be made concerning the above, decisions that require judgment and non-quantitative reasoning. What is an error, mistake, or any other occurrence that I should be concerned about. We often do not realize what we should be concerned about. · How do I detect an error or occurrence of interest? The previous bullet is often answered by—how extreme does my result have to be before I label it as non-random, e.g., 1 in 10, 1 in 1,000? When do I call the coin tosser or card player a cheat? How certain do I want to be that something unusual (cheating or other non-random error) is
Paper For Above instruction
Randomness is a fundamental concept in probability and statistics, underlying many processes and decisions in quality management. It refers to outcomes that are caused simply by chance, without any controlling influence or bias. Understanding what constitutes true randomness is essential for analyzing data, making predictions, and distinguishing between random variation and non-random errors in various processes.
At its core, randomness implies that each possible outcome has an equal probability of occurring. For example, in a fair coin toss, the chances of landing heads or tails are equal at 50%. This equality of probability ensures impartiality, making the outcome unpredictable and unbiased, which is essential for fair testing and sampling. However, true randomness can be elusive because many factors—such as imperfections in equipment, slight biases, or hidden influences—can introduce bias or distort outcomes, making them appear non-random even when they are, in a philosophical sense, unpredictable.
In practice, no process is perfectly random due to the influence of undetectable forces or imperfections—a concept supported by the idea that with enough information, future events could be predicted. Nonetheless, the real-world scenario involves complex interactions of forces that usually make outcomes sufficiently unpredictable for practical purposes. This unpredictability is what allows us to statistically analyze processes and outcomes, identifying when variations are due to chance versus when attributed to specific causes or errors.
Distinguishing random errors from non-random errors is a central challenge in quality management. Random errors, often called common causes, are inherent to the process and cannot be eliminated but can be reduced through precise measurement and process control. Non-random errors—or special causes—arise from identifiable, assignable factors that can often be eliminated or corrected once recognized. Statisticians utilize tools such as control charts and hypothesis testing to differentiate these errors, enabling effective decision-making.
For example, in manufacturing, fluctuations in product quality could be due to random variability—slight differences in raw materials, machine vibrations, or environmental conditions. Recognizing these as random helps avoid unnecessary adjustments or interventions that could disrupt stable processes. Conversely, a sudden shift in quality or a persistent defect indicates a non-random cause, prompting
investigation and correction.
Probability plays a crucial role in assessing the likelihood of observed outcomes. For example, flipping a coin five times and getting five heads occurs with a probability of 0.03% (1 in 32,768), which, if observed repeatedly, could be deemed unlikely enough to suggest bias or foul play. Conversely, certain unlikely outcomes, such as a specific six-number lottery win, might occur purely by chance, emphasizing the importance of statistical analysis in judgment calls about anomalies and irregularities.
The application of statistical tests enhances decision-making by quantifying the likelihood of events and their deviations from expectations based on probability models. For example, if a process produces defects less frequently than statistically expected, it could indicate an underlying issue; if more, then investigating whether a non-random cause exists. These analytical tools help managers avoid overreacting to normal fluctuations and focus on meaningful signals in data.
Deciding when an outcome is sufficiently unusual to warrant concern involves subjective judgment balanced with statistical evidence. For example, seeing five heads in a row prompts questions about fairness; however, knowing that such an event occurs with a probability of approximately 0.03% helps prevent mislabeling random variation as cheating. Similarly, rare successes or failures—like winning a lottery—are often just within the realm of chance, requiring careful probabilistic evaluation before assigning blame or suspicion.
In conclusion, understanding randomness, its implications, and how to differentiate between random and non-random variation are essential skills in quality management and statistical analysis. Recognizing that most variation in processes is due to either inherent randomness or assignable causes allows managers to make informed decisions, optimize processes, and avoid unnecessary interventions. Mastery of probability and proper application of statistical tools thus form the foundation for effective quality control, problem-solving, and decision-making in complex systems.
References
Montgomery, D. C. (2019). Introduction to Statistical Quality Control (8th ed.). Wiley.
Evans, J. R., & Lindsay, W. M. (2014). Managing for Quality and Performance Excellence (10th ed.). Cengage Learning.
Deming, W. E. (1986). Out of the Crisis. MIT Press.
Levine, D. M., Krehbiel, T. C., & Berenson, M. L. (2016). Statistics for Managers Using Microsoft Excel (8th ed.). Pearson.
Ross, S. M. (2014). Introduction to Probability and Statistics for Engineering and the Sciences (5th ed.). Academic Press.
Moore, D. S., & McCabe, G. P. (2017). Introduction to the Practice of Statistics (9th ed.). W. H. Freeman. Nelson, L. S. (2004). Statistical Methods in Quality Management. ASQ Quality Press.
Besterfield, D. H., et al. (2011). Quality Control (8th ed.). Pearson.
Ott, R. L., & Longnecker, M. (2015). An Introduction to Statistical Methods and Data Analysis. Cengage Learning.
Woodall, W. H. (2000). Dependence considerations in SPC monitoring schemes. Journal of Quality Technology, 32(2), 182-196.
