Software – The OM Blog by Heizer, Render, & Munson

OM in the News: The Robotics Supply Chain

The next 20 years are not just about making robots better, but also about how they will be used in all sorts of industries, from small tests to big factories. The real challenge is having specialized engineering skills, great manufacturing, and dominating software, reports Industry Week (March 11, 2026).

There are 6 key areas that make all the difference in this industry. Here is a breakdown of the cost of the parts that go into a robot:

1. Actuators & Gearboxes (35-40%): The physical muscle.

2. Robot Structure / Manipulators (15-20%): The physical frame and integration.

3. Sensors & Perception (10-15%): The eyes and ears.

4. AI Compute / Control (10-15%): The operational brain.

5. Battery / Power Systems (10-15%): The energy storage for mobile units.

6. Precision Motion Components (5-10%): The components required for fine movements.

This list shows that a robotics breakthrough isn’t just software advances; it depends on physical components and the supply chains that produce them. But there are 3 chokepoints (bottlenecks).

#1: Precision Reducers, controlled by Japan. Robots can’t move with a lot of power and precision without special parts (harmonic and cycloidal reducers). Two companies in Japan make 70% of these parts used all over the world. Spending more money won’t allow other companies to make these parts, because they need special knowledge about metals and years of experience making precise parts.

#2: AI Compute (The Intelligence Standard), controlled by the U.S. Today’s robots, especially those that use reinforcement learning, need powerful computers to work properly. NVIDIA’s CUDA system has become the leading platform used by robots that learn and think. Making a better chip is not enough if you can’t replace the software that all robotics engineers already use.

#3: Battery Supply Chain, controlled by China. Robots are changing from big, stationary machines to mobile ones. This means batteries are now a crucial part of making them work. One company in China, CATL, controls 1/3 of the world’s battery market. China has a very strong grip on this supply chain.

The global map of robotics is specialized. There is a multi-polar supply chain that is difficult to disrupt:

USA: “The “Brain.” (software, autonomy, AI compute).

Japan: The “Hardware King.” (motors, gearboxes, precision engineering).

Germany: The “Precision Engineer.” ( mechanical systems, high-end production).

China: The “Scale & Power.” (manufacturing speed, massive infrastructure, battery supremacy).

Taiwan: The “Linear Specialist.” ( The linear guides and ball screws essential for motion).

Classroom discussion questions:

Why must operations managers understand these costs and bottlenecks?
What are the supply chain implications?

Guest Post: Using Software in Your Operations Management Course

Prof. Howard Weiss’s suggestions may help you teach your OM course this semester.

As the developer of POM for Windows and Excel OM for the Heizer/Render/Munson text, I am naturally biased about the use of software in an OM class. In my opinion, students in an OM course, as opposed to an Operations Research course, should not be bogged down in the mechanics of computation but rather should understand the model concepts, inputs and outputs. There are four categories of problems that are strong candidates for assigning students to use POM, Excel OM or Excel.

Problems with a large data set There is no reason to ask students to solve PERT/CPM models with a large number of activities by hand. It makes much more sense to have the students enter the data and solve the problem using software, and then stress the concepts such as slack or crashing. In forecasting, students should understand the meaning and use of the trend and the error measures rather than spending time computing the intercept and slope. For control charts, if a student has computed results for 5 samples by hand does it really make sense to ask the student to compute the results by hand for 30 samples rather than using software to perform the computations and instead ask the student about his or her conclusions. There are many more examples of topics that include numerous computations – factor rating, center of gravity, assembly line balancing, ABC analysis, MRP, Lot-sizing and one machine sequencing come to mind. (Also, for models with much data, students should not spend a great deal of time entering data. It is very easy to copy and paste the data tables from MyOMLab to the software).

Iterative Models The discussion of LP, transportation, assignment and 2 machine scheduling, should focus on the formulation of the problems, the inputs, the goals and the interpretation of the results.

Models with multiple methods Some models have more than one approach. It is useful for the students to compare the results of these different methods without having to try every method by hand themselves. Obvious models are time- series analysis, assembly line balancing, aggregate planning and one-machine scheduling.

Models for which the software goes a step further than the text. In a few cases the software presents methods or results that are not part of the textbook. In one machine scheduling, Moore’s method is available to minimize the number of late jobs. Wagner-Whitin is included in the software for lot-sizing. Pairwise comparison is available to determine the layout with the minimum amount of movement.

OM in the News: Software Robots for Office Work

“Software robots have become one of the hottest fads in business automation, as a new wave of AI is poised to sweep through the back-office functions of large corporations,” writes The Financial Times (March 9, 2018). Software robots replicate the routine work humans often do in front of computer screens.

The rise of the bots promises to bring sweeping changes for cubicle dwellers. “Some 4 million in the U.S. are likely to see their jobs taken over by the end of 2021,” says one analyst. Each bot can handle the work it would take 3-4 full-time workers to perform. At $8,000-$9,000 a year in licensing costs, they are a lot cheaper.

But many automation experts say that the majority of these displaced workers will be kept on. Most companies will use the technology — known as “robotic process automation” — to automate only the most tedious aspects of back-office jobs, then retain the staff to work alongside the bots doing more interesting things. (There has been particular interest in Japan for the robots, given demographic trends that point to a coming shortage of workers).

Supporters think that applying machine learning to data could give companies important new insights into their operations. The automation software is also a natural place for bringing other types of AI into companies. Once a routine task is automated, it often makes sense to “plug in” an AI service from a third-party — for instance, using a vision system to analyze signatures as part of a bill-paying process. One insurance company, for example, has used bots to automate some of the work of 1,500 people who process new insurance requests, while also tapping into 3rd party “cognitive service” software to help make underwriting decisions.

Classroom discussion questions:

Do you agree with the industry analyst’s comment regarding the 4 million jobs?
Provide other examples of how office robots could make for higher productivity.

Guest Post: How We Use Software at Temple U.

Here is our 1st Guest Post. It comes from Prof. Howard Weiss at Temple U. As I mentioned in my Teaching Tip blog on 9/21, Howard has developed and maintains the software that accompanies our texts. He has been on the cutting edge of using computers to teach OM for over 2 decades.

Howard Weiss writes:

When I began using software in my OM class it was Lotus 123…remember that?! I would display the spreadsheets from the front of the classroom. But on their teaching evaluations, several students requested hands-on use of software rather than just watching me. I experimented by running half of each course in the classroom and half in the public computer lab. Recall that this was at the time when few students had PCs at home. The experiment was a success, and my departmental colleagues followed the half-time lab model. At the same time, in addition to public labs, teaching labs were being configured. Since then, roughly at the time when Windows and Excel began to become popular, all of our sections have been scheduled for half- time in the classroom and half-time in the teaching lab. We are a large school, so to make scheduling easier we have two sections of OM taught simultaneously. One section is in the classroom while the other is in the lab and vice-versa for the other lecture that week.

Changing to the alternating classroom/lab format meant a restructuring of the lectures. I try to lecture on qualitative material on the first day and then use the lab on the second day for the quantitative material rather than mixing the qualitative and quantitative material. My labs include exercises in Excel in order to build the students’ Excel skills, use of both POM for Windows and Excel OM so that the students can decide which they prefer, and use of the Active Models that accompany the Heizer/Render textbook.

Having developed POM for Windows and Excel OM, I spend a significant amount of my personal time continuously maintaining and improving these packages. I am always available to respond to any requests for help of any sort from you or your students. You can reach me at dsSoftware@prenhall.com.

If you would like to share your teaching experiences, please just email us and we will post your Guest Blog.

Teaching Tip:Using Software to Solve Homework Problems

When I look back at 35 years of textbook writing, I sometimes think that my biggest contribution to OM pedagogy was writing the early versions of AB:POM. These were the PC programs that allow students to solve about 95% of the homework problems in the text. I remember coding them in BASIC, in 1980 on my Apple 2e. Thank goodness for Professor Howard Weiss at Temple U., who had the skill to bring about the upgrade to POM for Windows (and then to create Excel OM) a decade or so later.

I usually spend a half hour demonstrating how to use these programs, free to students, the first week of class. They pick up on the inputs easily and are very happy to avoid doing regressions, SPC, and inventory modeling by hand. I know some profs don’t use software in their course, but my view has always been to stress output interpretation, rather than hand calculations.

Students have called me years later to tell me they still find the software useful in their jobs.

Which one to use, POM or Excel OM? I used to show both and let the students pick their favorite. But later I decided to focus on Excel OM, the Excel add-in. Its open coding shows the formula and looks like the spreadsheets they are used to. Any homework problems with an X next to it can be solved in Excel OM, and any with a P by POM (which is written in C++).