This paper looks at the design and implementation of a wireless data entry system using cheap and widely available personal digital assistants (PDAs) as a hardware platform. This problem is viewed through the context of a small transportation company that is transitioning to a medium-large scale operation and needs a custom tailored solution to improve payroll efficiency and reduce errors. In my report, I detail the design considerations, the user stories, and the pitfalls that come with implementing a new system that interfaces with existing applications.
This is a program I wrote to help high volume ebay sellers.
After struggling with 100+ orders per day on ebay, I realized I needed an automated system. Ebay and paypal are notoriously difficult to work with for large scale operations and the applications that interface with them tend to be short on features. With this in mind, I set out to create my own order processing system.
Before Paypal had an API, it was very difficult to get buyer information for creating internet based shipping labels. This program allowed you to log into your account and retrieve shipping information for your buyers. It also saved transaction snapshots, so you wouldn’t need an internet connection to view archived sales.
This application allowed you to keep in constant communication with your buyers to ensure they were happy by using a click and email system that notifies buyers of their shipment’s status. The notification idea is based on the system used by Dell for shipments where the customer is notified of shipment status through frequent, step-based email.
For example, a step 1 email would say order received, and would be marked as step 1 of 5. A step 2 email would say order confirmed and so on. For my order system, I used the following steps: order received, order processing, order processed, shipping label created, and order shipped. Using this method, customers were much more likely to leave positive feedback because I had kept in constant communication with them through a simple point and click interface.
Once Paypal created an API, I moved to Paypal Helper 2.0, which takes advantage of the new API to streamline large scale ebay order processing.
This project is an adaptation of an excel spreadsheet that was used for cost estimation. I added in a MapPoint control, so we wouldn’t need online access to work with maps. I also incorporated live Quickbooks data into this application, so that we could create formal quotes with prefilled fields on company letterhead with very little effort.
This application allows you to analyze different CPU scheduling algorithms. It currently uses 3 non preemptive algorithms (first in first out, shortest job next, non preemptive priority) and 3 preemptive algorithms (round robin, shortest time remaining, preemptive priority). I designed this application so that additional algorithms can be added in and analyzed alongside the built-in algorithms. Within the interface, you can generate random processes, create/edit your own processes, change the round robin time quantum, set all arrival times to 0, and save/load a set of processes.
Here is an early sketch I made for this application to help plan its implementation:
Early schematic for CPU scheduling program
Here is a screenshot of the working application:
CPU Scheduling Interface
The graph control used in this project is an open source C# control called ZedGraph.
