DataCollectionandAnalysis（资料收集与分析）.pptVIP

Data Collection and Analysis Data Collection and Analysis(資料收集與分析) By C. L. Hsieh Department of Industrial Management Aletheia University Introduction (介紹) “You can observe a lot just by watching” (你可以只以視覺方式來觀察許多數據) Data gathering results a conceptual model of how the system operated (資料收集可以產生一個觀念式模型來解釋系統是如何運作的) Data gathering should avoid ending up with lots of data but with very little useful information (資料收集應避免最後留下一堆資料但無太多有用的資訊) Questions for Data Gathering(資料收集的可能問題) What is the best procedure to follow? (資料收集的最佳程序為何?) What types of data should be gathered? (哪些資料種類應被收集?) What sources should be used ? (資料的來源為何?) What types of analyses should be performed on the data? (資料應進行何種分析?) How do you select the right probability distribution representing the data? (如何找出代表資料的分配?) How should data be documented? (資料如何文件化?) Guidelines for Data Gathering (資料收集應注意事項) Identify triggering events: (了解啟動活動原因) identify the causes or conditions that trigger the activities，e.g. the causes of downtime: failure, idle, unavailability of stock…… (了解啟動活動的原因或狀況，如停機原因：機器故障、閒置、缺貨…… ) Look for common grouping (分群以化簡資料) the solution is to reduce the data to common behaviors and patterns (化簡資料至一般行為與樣式) Identify general categories (確定一般性分類) Guidelines for Data Gathering (資料收集應注意事項) Focus on key impact factors (處理主要影響因子) Avoid little impact information (e.g. off-hour performance, extremely rare downtime, negligible move time..) (避免影響性小的因素，如加班特例、罕見的機器故障、可忽略的移動時間.) Separate input variables from response variables (區分輸入變數與回應變數) Input variables define how the system works (輸入變數決定系統運作) Response variables do not “drive” model behavior Guidelines for Data Gathering (資料收集應注意事項) Focus on essence rather than substance Capture cause-effect relationships and ignore meaningless details (著重因果關係，忽略細節) Focus on the activity of using resources or the delay of entity flow (system abstraction) (重視系統抽象層面) Isolate actual activity times. (確定真正活動時間) Exclude any extra time waiting (