Please enable JavaScript.
Coggle requires JavaScript to display documents.
Block 7: Management Science - Coggle Diagram
Block 7: Management Science
Management Science
branch of Operations Management (Taylor)
Quantitative Basis (data, models, algorithms)
Role of Managers
Previously to gather relevant data
Now, challenge is to identify which data is useful and how it can be used
Economic Order Quantity (EOQ)
Minimises Total Costs
Holding Costs
Order Costs
Inventory Control
Limitations
Assumes Steady Demand
Holding Cost may be Innaccurate
cost can be allocated precisely (activity-based costing) or apportion equally across inventory
Ignores External Dissruptions
Doesn't Consider Storage Limitations
Lean Operations and Just-in-Time
Link with Scientific Management
Minimum Human Intervention
Pre-defined Work Operations
Repetitive Short-cycle Work
Powerful First Line Supervisor
Conventional Managerial Hierarchy
JIT
Example: Toyota receive parts only when required
Why Avoid Inventories?
Customer Queues
waste customers' time and may upset/lose customers
requires area for waiting (or phone lines held for calls)
may put staff under pressure which reduce work quality
Physical Inventory
ties up working capital
high administrative and insurance cost
requires storage space
may become obsolete/deteriorate/damaged
Digital Information in Databases
cost of set-up, access, update and maintenance
requires memory capacity
may require secure or special environment
data may be corrupted/ lost/ obsolete
database need constant management
Limitations
if one step fails, it disrupt entire workflow
can make tasks monotonous and fail to sufficiently reward workers for their contributions to efficiency gain
Queueing Model
Little's Law
to calculate queue time and length by understanding how long customers spend in a system; factors can then be manipulated to reduce queue length/time
can help calculate optimal number of staff needed by linking throughput time with number of customers beings served, improving staffing efficiency
Limitations (apply for Little's Law)
Potential variability causes formulae used to understand queueing extremely complicated
Variability in rate at which customeers join a queue (Example: more people go to post office at lunch time, and less busy in the morning)
Variability in how long it takes to process each customer (Example: some people only need to post a letter which is quick, while some need to post 4 parcels which take longer)
Unintentional inaccurate inputs reduce reliability of model outputs
Some problems are unavoidable (ex: unexpected demand exceeds space)
Overemphasis on quantitative data and lack of focus on customer sentiment (qualitative)
Example: people may tolerate long queue at Disneyworld, but in other contexts like bank, even short waits lead to frustration
Thus, first-hand observation can offer valuable qualitative insights (ex: customer frustration, impact of physical space)
Queueing Information
Arrival Rate
Processing Rate
Expected Waiting Time in Queue
How Queueing Theory Can Be Used to Improve Firm Practice
Manipulating Factors to Reduce Queue Length
Example: encouraging people to move towards online banking
Reduce Demand Variability
Example: holiday companies charge higher during holidays but reduce price during off-peak periods
Planning Traffic Flows
Example: streaming customers by separating quick task from more complex interactions (reduce variability)
Reducing Frustrations at Peak Times
Example: meeting a Disney character while queueing ride
Using Queue to Build Sales
Example: strategically putting small tempting items like sweet near checkout lines
Big Data
Example: Amazon uses big data to recommend products based on customer browsing history
Evolution of Big Data
Information Revolution (1985)
World Wide Web (WWW)
Big Data expands through more sophisticated analysis, but creates job security concerns
Features
Volume
Variety
Velocity
Veracity
Benefits
Driving Innovation & Gaining Competitive Advantage
Improving Productivity
Internet of Things (IoT)
Example: GPS in delivery trucks provide business with real-time insights to improve services like predicting delivery times
Workplace Wearables & RFID (radio frequency identification)
RFID
data errors are almost eliminated
real-time info is available
important for both EOQ and lean management (automates stock tracking and reduces manual errors)
Workplace Wearables
updated approach of Taylor's time and motion studies
Example: Xerox analyse movement and interaction to understand where time is spent at work
Advantages of Workplace Wearables
Improve processes by understanding them more clearly
Track goods in transport, through trackers on vehicles or people
Automated data about staff attendance and hours
Disadvantages of Workplace Wearables
Staff may dislike being monitored; may cause stress; Example: Amazon warehouse workers have faced high stress due to constant productivity monitoring to meet strictly efficiency targets
Managing large amounts of data can be challenging, and there's risk of focusing on wrong metrics
Focusing too much on internal processes may cause business to overlook important external changes
Examples where Workers may be Glad to Make Use of Wearables
Exoskeletons suits to lift heavy materials safely
Smart glasses help workers quickly locate and pick items
Smartwatches or headsets to streamline communication
Artificial Intelligence (AI)
Technologies Used in AI
Natural Language Processing (NLP); Ex: virtual assistants like Siri, chatbots
Computer Vision; Ex: used in self-driving cars, wildlife monitoring
Generative AI; Ex: ChatGPT
Advantages of AI
Reduce human errors
Automates repetitive tasks
Efficiently processes large amounts of data
Limitations of AI
Increase risks of scams through voice recognition technology
Algorithm bias can reinforce discrimination
May displace human workers
Lack of transparency in AI decision-making
High implementation costs
AI-powered surveillance threatens privacy and civil liberties
regulation manage risks, excessive rules may slow innovation
May develop in unexpected ways, especially when trained on unstructured data (as it does not follow a predefined format), which may lead to problems
Example: Chatbots on e-commerce websites answer customer questions without human involvement